Multilevel Effects of Leader Charisma on Follower Satisfaction

Permanent Link: http://ufdc.ufl.edu/UFE0011346/00001

Material Information

Title: Multilevel Effects of Leader Charisma on Follower Satisfaction
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0011346:00001

Permanent Link: http://ufdc.ufl.edu/UFE0011346/00001

Material Information

Title: Multilevel Effects of Leader Charisma on Follower Satisfaction
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0011346:00001

This item has the following downloads:

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101129_AAAABG INGEST_TIME 2010-11-30T02:35:55Z PACKAGE UFE0011346_00001
1053954 F20101129_AAAWDI halverson_k_Page_43.tif
58453 F20101129_AAAVWL halverson_k_Page_26.jpg
107125 F20101129_AAAWIG UFE0011346_00001.xml FULL
F20101129_AAAWDJ halverson_k_Page_45.tif
5013 F20101129_AAAWIH halverson_k_Page_02.QC.jpg
F20101129_AAAWDK halverson_k_Page_46.tif
74485 F20101129_AAAVWM halverson_k_Page_28.jpg
1901 F20101129_AAAWII halverson_k_Page_02thm.jpg
F20101129_AAAWDL halverson_k_Page_47.tif
67367 F20101129_AAAVWN halverson_k_Page_29.jpg
4016 F20101129_AAAWIJ halverson_k_Page_03.QC.jpg
25271604 F20101129_AAAWDM halverson_k_Page_48.tif
76080 F20101129_AAAVWO halverson_k_Page_30.jpg
7238 F20101129_AAAWIK halverson_k_Page_05.QC.jpg
F20101129_AAAWDN halverson_k_Page_49.tif
73025 F20101129_AAAVWP halverson_k_Page_31.jpg
2462 F20101129_AAAWIL halverson_k_Page_05thm.jpg
F20101129_AAAWDO halverson_k_Page_50.tif
71175 F20101129_AAAVWQ halverson_k_Page_32.jpg
3331 F20101129_AAAWIM halverson_k_Page_07thm.jpg
F20101129_AAAWDP halverson_k_Page_51.tif
71869 F20101129_AAAVWR halverson_k_Page_33.jpg
3912 F20101129_AAAWIN halverson_k_Page_08thm.jpg
F20101129_AAAWDQ halverson_k_Page_52.tif
74773 F20101129_AAAVWS halverson_k_Page_34.jpg
6377 F20101129_AAAWIO halverson_k_Page_09.QC.jpg
F20101129_AAAWDR halverson_k_Page_53.tif
69751 F20101129_AAAVWT halverson_k_Page_36.jpg
2147 F20101129_AAAWIP halverson_k_Page_09thm.jpg
F20101129_AAAWDS halverson_k_Page_54.tif
71509 F20101129_AAAVWU halverson_k_Page_37.jpg
18603 F20101129_AAAWIQ halverson_k_Page_10.QC.jpg
F20101129_AAAWDT halverson_k_Page_55.tif
66528 F20101129_AAAVWV halverson_k_Page_38.jpg
5196 F20101129_AAAWIR halverson_k_Page_10thm.jpg
F20101129_AAAWDU halverson_k_Page_56.tif
75140 F20101129_AAAVWW halverson_k_Page_39.jpg
2358 F20101129_AAAWIS halverson_k_Page_11thm.jpg
F20101129_AAAWDV halverson_k_Page_57.tif
72713 F20101129_AAAVWX halverson_k_Page_40.jpg
6305 F20101129_AAAWIT halverson_k_Page_13thm.jpg
F20101129_AAAWDW halverson_k_Page_58.tif
6763 F20101129_AAAVUA halverson_k_Page_91thm.jpg
69207 F20101129_AAAVWY halverson_k_Page_41.jpg
6567 F20101129_AAAWIU halverson_k_Page_14thm.jpg
F20101129_AAAWDX halverson_k_Page_59.tif
108680 F20101129_AAAVUB halverson_k_Page_13.jp2
70956 F20101129_AAAVWZ halverson_k_Page_43.jpg
7320 F20101129_AAAWIV halverson_k_Page_16.QC.jpg
22292 F20101129_AAAVUC halverson_k_Page_23.QC.jpg
5836 F20101129_AAAWIW halverson_k_Page_17thm.jpg
91267 F20101129_AAAWBA halverson_k_Page_71.jp2
F20101129_AAAWDY halverson_k_Page_60.tif
19403 F20101129_AAAVUD halverson_k_Page_65.QC.jpg
98759 F20101129_AAAVZA halverson_k_Page_04.jp2
6664 F20101129_AAAWIX halverson_k_Page_18thm.jpg
103572 F20101129_AAAWBB halverson_k_Page_72.jp2
F20101129_AAAWDZ halverson_k_Page_61.tif
107700 F20101129_AAAVUE halverson_k_Page_36.jp2
25292 F20101129_AAAVZB halverson_k_Page_05.jp2
22464 F20101129_AAAWIY halverson_k_Page_19.QC.jpg
67921 F20101129_AAAWBC halverson_k_Page_73.jp2
73279 F20101129_AAAVUF halverson_k_Page_14.jpg
1051980 F20101129_AAAVZC halverson_k_Page_06.jp2
23324 F20101129_AAAWIZ halverson_k_Page_21.QC.jpg
22695 F20101129_AAAVUG halverson_k_Page_13.QC.jpg
6435 F20101129_AAAWGA halverson_k_Page_78thm.jpg
1051976 F20101129_AAAVZD halverson_k_Page_07.jp2
781936 F20101129_AAAWBD halverson_k_Page_74.jp2
F20101129_AAAVUH halverson_k_Page_95.tif
23275 F20101129_AAAWGB halverson_k_Page_32.QC.jpg
1051975 F20101129_AAAVZE halverson_k_Page_08.jp2
88270 F20101129_AAAWBE halverson_k_Page_75.jp2
22800 F20101129_AAAVUI halverson_k_Page_60.QC.jpg
6721 F20101129_AAAWGC halverson_k_Page_53thm.jpg
85985 F20101129_AAAVZF halverson_k_Page_10.jp2
95208 F20101129_AAAWBF halverson_k_Page_76.jp2
23790 F20101129_AAAWLA halverson_k_Page_91.QC.jpg
F20101129_AAAVUJ halverson_k_Page_03.tif
7025 F20101129_AAAWGD halverson_k_Page_01.QC.jpg
25602 F20101129_AAAVZG halverson_k_Page_11.jp2
113237 F20101129_AAAWBG halverson_k_Page_77.jp2
7033 F20101129_AAAWLB halverson_k_Page_92thm.jpg
20805 F20101129_AAAWGE halverson_k_Page_71.QC.jpg
96059 F20101129_AAAVZH halverson_k_Page_12.jp2
105713 F20101129_AAAWBH halverson_k_Page_78.jp2
6837 F20101129_AAAWLC halverson_k_Page_93thm.jpg
6414 F20101129_AAAVUK halverson_k_Page_19thm.jpg
6484 F20101129_AAAWGF halverson_k_Page_20thm.jpg
111118 F20101129_AAAVZI halverson_k_Page_14.jp2
107752 F20101129_AAAWBI halverson_k_Page_79.jp2
6801 F20101129_AAAWLD halverson_k_Page_94thm.jpg
132595 F20101129_AAAVUL halverson_k_Page_95.jp2
5586 F20101129_AAAWGG halverson_k_Page_88thm.jpg
111418 F20101129_AAAVZJ halverson_k_Page_15.jp2
105578 F20101129_AAAWBJ halverson_k_Page_81.jp2
21844 F20101129_AAAWLE halverson_k_Page_96.QC.jpg
F20101129_AAAVUM halverson_k_Page_82.tif
25179 F20101129_AAAWGH halverson_k_Page_93.QC.jpg
27400 F20101129_AAAVZK halverson_k_Page_16.jp2
109070 F20101129_AAAWBK halverson_k_Page_82.jp2
14102 F20101129_AAAWLF halverson_k_Page_97.QC.jpg
F20101129_AAAVUN halverson_k_Page_75.tif
2373 F20101129_AAAWGI halverson_k_Page_16thm.jpg
97955 F20101129_AAAVZL halverson_k_Page_17.jp2
112135 F20101129_AAAWBL halverson_k_Page_83.jp2
94399 F20101129_AAAVUO halverson_k_Page_62.jp2
5957 F20101129_AAAWGJ halverson_k_Page_67thm.jpg
115036 F20101129_AAAVZM halverson_k_Page_18.jp2
111025 F20101129_AAAWBM halverson_k_Page_84.jp2
821201 F20101129_AAAVUP halverson_k.pdf
6399 F20101129_AAAWGK halverson_k_Page_43thm.jpg
106927 F20101129_AAAVZN halverson_k_Page_19.jp2
108741 F20101129_AAAWBN halverson_k_Page_85.jp2
6539 F20101129_AAAVUQ halverson_k_Page_60thm.jpg
23794 F20101129_AAAWGL halverson_k_Page_45.QC.jpg
112307 F20101129_AAAVZO halverson_k_Page_20.jp2
114347 F20101129_AAAWBO halverson_k_Page_86.jp2
74793 F20101129_AAAVUR halverson_k_Page_59.jpg
6495 F20101129_AAAWGM halverson_k_Page_21thm.jpg
98016 F20101129_AAAWBP halverson_k_Page_88.jp2
23934 F20101129_AAAVUS halverson_k_Page_20.QC.jpg
1658 F20101129_AAAWGN halverson_k_Page_03thm.jpg
109338 F20101129_AAAVZP halverson_k_Page_21.jp2
126088 F20101129_AAAWBQ halverson_k_Page_89.jp2
18330 F20101129_AAAVUT halverson_k_Page_74.QC.jpg
12079 F20101129_AAAWGO halverson_k_Page_07.QC.jpg
110480 F20101129_AAAVZQ halverson_k_Page_22.jp2
132504 F20101129_AAAWBR halverson_k_Page_90.jp2
17185 F20101129_AAAVUU halverson_k_Page_25.QC.jpg
24210 F20101129_AAAWGP halverson_k_Page_52.QC.jpg
104951 F20101129_AAAVZR halverson_k_Page_23.jp2
121857 F20101129_AAAWBS halverson_k_Page_91.jp2
F20101129_AAAVUV halverson_k_Page_14.tif
6811 F20101129_AAAWGQ halverson_k_Page_55thm.jpg
76482 F20101129_AAAVZS halverson_k_Page_25.jp2
1051951 F20101129_AAAWBT halverson_k_Page_92.jp2
72384 F20101129_AAAVUW halverson_k_Page_24.jpg
22516 F20101129_AAAWGR halverson_k_Page_41.QC.jpg
903481 F20101129_AAAVZT halverson_k_Page_26.jp2
126697 F20101129_AAAWBU halverson_k_Page_93.jp2
72544 F20101129_AAAVUX halverson_k_Page_22.jpg
14338 F20101129_AAAWGS halverson_k_Page_08.QC.jpg
114076 F20101129_AAAVZU halverson_k_Page_28.jp2
127423 F20101129_AAAWBV halverson_k_Page_94.jp2
6208 F20101129_AAAWGT halverson_k_Page_61thm.jpg
114724 F20101129_AAAVZV halverson_k_Page_30.jp2
20093 F20101129_AAAVUY halverson_k_Page_67.QC.jpg
22908 F20101129_AAAWGU halverson_k_Page_37.QC.jpg
112326 F20101129_AAAVZW halverson_k_Page_31.jp2
110414 F20101129_AAAWBW halverson_k_Page_96.jp2
21989 F20101129_AAAVUZ halverson_k_Page_78.QC.jpg
26191 F20101129_AAAWGV halverson_k_Page_95.QC.jpg
110201 F20101129_AAAVZX halverson_k_Page_33.jp2
60657 F20101129_AAAWBX halverson_k_Page_97.jp2
22822 F20101129_AAAWGW halverson_k_Page_49.QC.jpg
70665 F20101129_AAAVXA halverson_k_Page_44.jpg
113982 F20101129_AAAVZY halverson_k_Page_34.jp2
F20101129_AAAWBY halverson_k_Page_01.tif
6593 F20101129_AAAWGX halverson_k_Page_49thm.jpg
73002 F20101129_AAAVXB halverson_k_Page_45.jpg
109086 F20101129_AAAVZZ halverson_k_Page_37.jp2
F20101129_AAAWBZ halverson_k_Page_02.tif
5615 F20101129_AAAWGY halverson_k_Page_12thm.jpg
77201 F20101129_AAAVXC halverson_k_Page_47.jpg
6899 F20101129_AAAWGZ halverson_k_Page_95thm.jpg
62640 F20101129_AAAVXD halverson_k_Page_48.jpg
F20101129_AAAWEA halverson_k_Page_62.tif
70916 F20101129_AAAVXE halverson_k_Page_49.jpg
F20101129_AAAWEB halverson_k_Page_63.tif
70903 F20101129_AAAVXF halverson_k_Page_50.jpg
F20101129_AAAWEC halverson_k_Page_64.tif
6239 F20101129_AAAWJA halverson_k_Page_23thm.jpg
73283 F20101129_AAAVXG halverson_k_Page_51.jpg
F20101129_AAAWED halverson_k_Page_65.tif
23257 F20101129_AAAWJB halverson_k_Page_24.QC.jpg
73876 F20101129_AAAVXH halverson_k_Page_52.jpg
F20101129_AAAWEE halverson_k_Page_66.tif
4932 F20101129_AAAWJC halverson_k_Page_25thm.jpg
73994 F20101129_AAAVXI halverson_k_Page_53.jpg
6579 F20101129_AAAVSK halverson_k_Page_51thm.jpg
F20101129_AAAWEF halverson_k_Page_67.tif
6674 F20101129_AAAWJD halverson_k_Page_27thm.jpg
73131 F20101129_AAAVXJ halverson_k_Page_54.jpg
100229 F20101129_AAAVSL halverson_k_Page_38.jp2
F20101129_AAAWEG halverson_k_Page_68.tif
22003 F20101129_AAAWJE halverson_k_Page_29.QC.jpg
75917 F20101129_AAAVXK halverson_k_Page_55.jpg
F20101129_AAAVSM halverson_k_Page_87.tif
F20101129_AAAWEH halverson_k_Page_69.tif
6513 F20101129_AAAWJF halverson_k_Page_31thm.jpg
69187 F20101129_AAAVXL halverson_k_Page_56.jpg
23197 F20101129_AAAVSN halverson_k_Page_79.QC.jpg
F20101129_AAAWEI halverson_k_Page_70.tif
6324 F20101129_AAAWJG halverson_k_Page_32thm.jpg
32270 F20101129_AAAVXM halverson_k_Page_57.jpg
24262 F20101129_AAAVSO halverson_k_Page_28.QC.jpg
F20101129_AAAWEJ halverson_k_Page_71.tif
24532 F20101129_AAAWJH halverson_k_Page_34.QC.jpg
108392 F20101129_AAAVSP halverson_k_Page_24.jp2
F20101129_AAAWEK halverson_k_Page_72.tif
6606 F20101129_AAAWJI halverson_k_Page_35thm.jpg
56982 F20101129_AAAVXN halverson_k_Page_58.jpg
F20101129_AAAVSQ halverson_k_Page_44.tif
F20101129_AAAWEL halverson_k_Page_74.tif
22635 F20101129_AAAWJJ halverson_k_Page_36.QC.jpg
70904 F20101129_AAAVXO halverson_k_Page_61.jpg
6404 F20101129_AAAVSR halverson_k_Page_79thm.jpg
F20101129_AAAWEM halverson_k_Page_76.tif
F20101129_AAAWJK halverson_k_Page_36thm.jpg
61628 F20101129_AAAVXP halverson_k_Page_62.jpg
55052 F20101129_AAAVSS halverson_k_Page_87.jp2
F20101129_AAAWEN halverson_k_Page_78.tif
24801 F20101129_AAAWJL halverson_k_Page_39.QC.jpg
61403 F20101129_AAAVXQ halverson_k_Page_63.jpg
F20101129_AAAVST halverson_k_Page_85.tif
F20101129_AAAWEO halverson_k_Page_79.tif
23561 F20101129_AAAWJM halverson_k_Page_40.QC.jpg
64943 F20101129_AAAVXR halverson_k_Page_64.jpg
104014 F20101129_AAAVSU halverson_k_Page_56.jp2
F20101129_AAAWEP halverson_k_Page_81.tif
6577 F20101129_AAAWJN halverson_k_Page_42thm.jpg
58093 F20101129_AAAVXS halverson_k_Page_65.jpg
111701 F20101129_AAAVSV halverson_k_Page_54.jp2
F20101129_AAAWEQ halverson_k_Page_83.tif
22962 F20101129_AAAWJO halverson_k_Page_44.QC.jpg
57319 F20101129_AAAVXT halverson_k_Page_66.jpg
72103 F20101129_AAAVSW halverson_k_Page_42.jpg
F20101129_AAAWER halverson_k_Page_84.tif
F20101129_AAAWJP halverson_k_Page_45thm.jpg
88130 F20101129_AAAVSX halverson_k_Page_94.jpg
F20101129_AAAWES halverson_k_Page_86.tif
63355 F20101129_AAAVXU halverson_k_Page_67.jpg
22236 F20101129_AAAWJQ halverson_k_Page_46.QC.jpg
20663 F20101129_AAAVSY halverson_k_Page_12.QC.jpg
F20101129_AAAWET halverson_k_Page_88.tif
17402 F20101129_AAAVXV halverson_k_Page_68.jpg
6231 F20101129_AAAWJR halverson_k_Page_46thm.jpg
20508 F20101129_AAAVSZ halverson_k_Page_63.QC.jpg
F20101129_AAAWEU halverson_k_Page_90.tif
55773 F20101129_AAAVXW halverson_k_Page_69.jpg
24961 F20101129_AAAWJS halverson_k_Page_47.QC.jpg
F20101129_AAAWEV halverson_k_Page_91.tif
71721 F20101129_AAAVXX halverson_k_Page_70.jpg
19738 F20101129_AAAWJT halverson_k_Page_48.QC.jpg
F20101129_AAAWEW halverson_k_Page_92.tif
13938 F20101129_AAAVVA halverson_k_Page_03.jpg
61946 F20101129_AAAVXY halverson_k_Page_71.jpg
5816 F20101129_AAAWJU halverson_k_Page_48thm.jpg
F20101129_AAAWEX halverson_k_Page_93.tif
74130 F20101129_AAAVVB halverson_k_Page_27.jpg
69128 F20101129_AAAVXZ halverson_k_Page_72.jpg
23178 F20101129_AAAWJV halverson_k_Page_50.QC.jpg
F20101129_AAAWEY halverson_k_Page_94.tif
23496 F20101129_AAAVVC halverson_k_Page_42.QC.jpg
6403 F20101129_AAAWJW halverson_k_Page_50thm.jpg
F20101129_AAAWCA halverson_k_Page_05.tif
24633 F20101129_AAAVVD halverson_k_Page_86.QC.jpg
24059 F20101129_AAAWJX halverson_k_Page_51.QC.jpg
F20101129_AAAWCB halverson_k_Page_06.tif
F20101129_AAAWEZ halverson_k_Page_96.tif
6542 F20101129_AAAVVE halverson_k_Page_82thm.jpg
6605 F20101129_AAAWJY halverson_k_Page_52thm.jpg
F20101129_AAAWCC halverson_k_Page_07.tif
F20101129_AAAVVF halverson_k_Page_04.tif
24551 F20101129_AAAWJZ halverson_k_Page_53.QC.jpg
F20101129_AAAWCD halverson_k_Page_08.tif
6421 F20101129_AAAVVG halverson_k_Page_59thm.jpg
22712 F20101129_AAAWHA halverson_k_Page_59.QC.jpg
F20101129_AAAWCE halverson_k_Page_09.tif
F20101129_AAAVVH halverson_k_Page_32.jp2
6436 F20101129_AAAWHB halverson_k_Page_37thm.jpg
F20101129_AAAWCF halverson_k_Page_10.tif
24382 F20101129_AAAVVI halverson_k_Page_27.QC.jpg
6543 F20101129_AAAWHC halverson_k_Page_84thm.jpg
F20101129_AAAWCG halverson_k_Page_12.tif
112303 F20101129_AAAVVJ halverson_k_Page_35.jp2
22666 F20101129_AAAWHD halverson_k_Page_61.QC.jpg
F20101129_AAAWCH halverson_k_Page_13.tif
840005 F20101129_AAAVVK halverson_k_Page_64.jp2
5901 F20101129_AAAWHE halverson_k_Page_96thm.jpg
F20101129_AAAWCI halverson_k_Page_15.tif
23485 F20101129_AAAWHF halverson_k_Page_33.QC.jpg
F20101129_AAAWCJ halverson_k_Page_16.tif
71433 F20101129_AAAVVL halverson_k_Page_60.jpg
6190 F20101129_AAAWHG halverson_k_Page_29thm.jpg
F20101129_AAAWCK halverson_k_Page_17.tif
86836 F20101129_AAAVVM halverson_k_Page_65.jp2
24459 F20101129_AAAWHH halverson_k_Page_30.QC.jpg
F20101129_AAAWCL halverson_k_Page_18.tif
74044 F20101129_AAAVVN halverson_k_Page_35.jpg
21268 F20101129_AAAWHI halverson_k_Page_38.QC.jpg
F20101129_AAAWCM halverson_k_Page_19.tif
25444 F20101129_AAAVVO halverson_k_Page_92.QC.jpg
23423 F20101129_AAAWHJ halverson_k_Page_22.QC.jpg
F20101129_AAAWCN halverson_k_Page_20.tif
73526 F20101129_AAAVVP UFE0011346_00001.mets
23021 F20101129_AAAWHK halverson_k_Page_85.QC.jpg
F20101129_AAAWCO halverson_k_Page_21.tif
21763 F20101129_AAAWHL halverson_k_Page_04.QC.jpg
F20101129_AAAWCP halverson_k_Page_22.tif
F20101129_AAAWHM halverson_k_Page_11.QC.jpg
F20101129_AAAWCQ halverson_k_Page_24.tif
22276 F20101129_AAAVVS halverson_k_Page_01.jpg
5966 F20101129_AAAWHN halverson_k_Page_68.QC.jpg
F20101129_AAAWCR halverson_k_Page_25.tif
15965 F20101129_AAAVVT halverson_k_Page_02.jpg
25189 F20101129_AAAWHO halverson_k_Page_94.QC.jpg
F20101129_AAAWCS halverson_k_Page_26.tif
66668 F20101129_AAAVVU halverson_k_Page_04.jpg
6621 F20101129_AAAWHP halverson_k_Page_28thm.jpg
F20101129_AAAWCT halverson_k_Page_27.tif
21455 F20101129_AAAVVV halverson_k_Page_05.jpg
6518 F20101129_AAAWHQ halverson_k_Page_70thm.jpg
F20101129_AAAWCU halverson_k_Page_28.tif
67439 F20101129_AAAVVW halverson_k_Page_06.jpg
6524 F20101129_AAAWHR halverson_k_Page_80thm.jpg
F20101129_AAAWCV halverson_k_Page_29.tif
45266 F20101129_AAAVVX halverson_k_Page_07.jpg
6870 F20101129_AAAWHS halverson_k_Page_86thm.jpg
F20101129_AAAWCW halverson_k_Page_30.tif
23631 F20101129_AAAVTA halverson_k_Page_84.QC.jpg
49755 F20101129_AAAVVY halverson_k_Page_08.jpg
5960 F20101129_AAAWHT halverson_k_Page_64thm.jpg
106292 F20101129_AAAVTB halverson_k_Page_80.jp2
20748 F20101129_AAAVVZ halverson_k_Page_09.jpg
24022 F20101129_AAAWHU halverson_k_Page_31.QC.jpg
F20101129_AAAVTC halverson_k_Page_32.tif
5764 F20101129_AAAWHV halverson_k_Page_75thm.jpg
F20101129_AAAWCX halverson_k_Page_31.tif
5524 F20101129_AAAWHW halverson_k_Page_66thm.jpg
F20101129_AAAVTD halverson_k_Page_89.tif
47320 F20101129_AAAVYA halverson_k_Page_73.jpg
114733 F20101129_AAAWAA halverson_k_Page_39.jp2
F20101129_AAAWCY halverson_k_Page_33.tif
5427 F20101129_AAAWHX halverson_k_Page_58thm.jpg
59339 F20101129_AAAVYB halverson_k_Page_74.jpg
111338 F20101129_AAAWAB halverson_k_Page_40.jp2
F20101129_AAAWCZ halverson_k_Page_34.tif
114663 F20101129_AAAVTE halverson_k_Page_27.jp2
4827 F20101129_AAAWHY halverson_k_Page_73thm.jpg
61434 F20101129_AAAVYC halverson_k_Page_75.jpg
106470 F20101129_AAAWAC halverson_k_Page_41.jp2
24266 F20101129_AAAVTF halverson_k_Page_35.QC.jpg
6835 F20101129_AAAWHZ halverson_k_Page_47thm.jpg
2400 F20101129_AAAWFA halverson_k_Page_01thm.jpg
64111 F20101129_AAAVYD halverson_k_Page_76.jpg
111550 F20101129_AAAWAD halverson_k_Page_42.jp2
F20101129_AAAVTG halverson_k_Page_97.tif
24831 F20101129_AAAWFB halverson_k_Page_55.QC.jpg
74395 F20101129_AAAVYE halverson_k_Page_77.jpg
106960 F20101129_AAAWAE halverson_k_Page_43.jp2
19668 F20101129_AAAVTH halverson_k_Page_75.QC.jpg
18162 F20101129_AAAWFC halverson_k_Page_66.QC.jpg
69640 F20101129_AAAVYF halverson_k_Page_78.jpg
108590 F20101129_AAAWAF halverson_k_Page_44.jp2
71858 F20101129_AAAVTI halverson_k_Page_85.jpg
23736 F20101129_AAAWKA halverson_k_Page_54.QC.jpg
12795 F20101129_AAAWFD halverson_k_Page_87.QC.jpg
71366 F20101129_AAAVYG halverson_k_Page_79.jpg
112420 F20101129_AAAWAG halverson_k_Page_45.jp2
6555 F20101129_AAAWKB halverson_k_Page_54thm.jpg
F20101129_AAAWFE halverson_k_Page_22thm.jpg
70908 F20101129_AAAVYH halverson_k_Page_80.jpg
103142 F20101129_AAAWAH halverson_k_Page_46.jp2
70183 F20101129_AAAVTJ halverson_k_Page_19.jpg
22131 F20101129_AAAWKC halverson_k_Page_56.QC.jpg
2115 F20101129_AAAWFF halverson_k_Page_68thm.jpg
70142 F20101129_AAAVYI halverson_k_Page_81.jpg
116246 F20101129_AAAWAI halverson_k_Page_47.jp2
6368 F20101129_AAAVTK halverson_k_Page_38thm.jpg
6319 F20101129_AAAWKD halverson_k_Page_56thm.jpg
6315 F20101129_AAAWFG halverson_k_Page_41thm.jpg
71163 F20101129_AAAVYJ halverson_k_Page_82.jpg
108419 F20101129_AAAWAJ halverson_k_Page_49.jp2
3738 F20101129_AAAVTL halverson_k_Page_87thm.jpg
10445 F20101129_AAAWKE halverson_k_Page_57.QC.jpg
17366 F20101129_AAAWFH halverson_k_Page_06.QC.jpg
73906 F20101129_AAAVYK halverson_k_Page_83.jpg
108905 F20101129_AAAWAK halverson_k_Page_50.jp2
F20101129_AAAVTM halverson_k_Page_77.tif
18380 F20101129_AAAWKF halverson_k_Page_58.QC.jpg
6505 F20101129_AAAWFI halverson_k_Page_24thm.jpg
72333 F20101129_AAAVYL halverson_k_Page_84.jpg
111456 F20101129_AAAWAL halverson_k_Page_51.jp2
64794 F20101129_AAAVTN halverson_k_Page_17.jpg
5914 F20101129_AAAWKG halverson_k_Page_62thm.jpg
6111 F20101129_AAAWFJ halverson_k_Page_04thm.jpg
74778 F20101129_AAAVYM halverson_k_Page_86.jpg
113785 F20101129_AAAWAM halverson_k_Page_52.jp2
102264 F20101129_AAAVTO halverson_k_Page_29.jp2
5521 F20101129_AAAWKH halverson_k_Page_63thm.jpg
6282 F20101129_AAAWFK halverson_k_Page_44thm.jpg
39024 F20101129_AAAVYN halverson_k_Page_87.jpg
114498 F20101129_AAAWAN halverson_k_Page_53.jp2
370646 F20101129_AAAVTP halverson_k_Page_09.jp2
5655 F20101129_AAAWKI halverson_k_Page_65thm.jpg
4139 F20101129_AAAWFL halverson_k_Page_97thm.jpg
114915 F20101129_AAAWAO halverson_k_Page_55.jp2
F20101129_AAAVTQ halverson_k_Page_73.tif
18037 F20101129_AAAWKJ halverson_k_Page_69.QC.jpg
23773 F20101129_AAAWFM halverson_k_Page_15.QC.jpg
70824 F20101129_AAAVYO halverson_k_Page_88.jpg
44438 F20101129_AAAWAP halverson_k_Page_57.jp2
67128 F20101129_AAAVTR halverson_k_Page_46.jpg
5148 F20101129_AAAWKK halverson_k_Page_69thm.jpg
20665 F20101129_AAAWFN halverson_k_Page_64.QC.jpg
87948 F20101129_AAAVYP halverson_k_Page_89.jpg
84948 F20101129_AAAWAQ halverson_k_Page_58.jp2
6482 F20101129_AAAVTS halverson_k_Page_15thm.jpg
22126 F20101129_AAAWKL halverson_k_Page_72.QC.jpg
4622 F20101129_AAAWFO halverson_k_Page_06thm.jpg
87825 F20101129_AAAVYQ halverson_k_Page_90.jpg
1051922 F20101129_AAAWAR halverson_k_Page_59.jp2
F20101129_AAAVTT halverson_k_Page_80.tif
6265 F20101129_AAAWKM halverson_k_Page_72thm.jpg
20290 F20101129_AAAWFP halverson_k_Page_88.QC.jpg
79556 F20101129_AAAVYR halverson_k_Page_91.jpg
108762 F20101129_AAAWAS halverson_k_Page_60.jp2
6722 F20101129_AAAVTU halverson_k_Page_39thm.jpg
15850 F20101129_AAAWKN halverson_k_Page_73.QC.jpg
5984 F20101129_AAAWFQ halverson_k_Page_71thm.jpg
93999 F20101129_AAAVYS halverson_k_Page_92.jpg
107866 F20101129_AAAWAT halverson_k_Page_61.jp2
918971 F20101129_AAAVTV halverson_k_Page_48.jp2
5377 F20101129_AAAWKO halverson_k_Page_74thm.jpg
6724 F20101129_AAAWFR halverson_k_Page_30thm.jpg
88642 F20101129_AAAVYT halverson_k_Page_93.jpg
89660 F20101129_AAAWAU halverson_k_Page_63.jp2
20992 F20101129_AAAVTW halverson_k_Page_17.QC.jpg
20275 F20101129_AAAWKP halverson_k_Page_76.QC.jpg
5667 F20101129_AAAWFS halverson_k_Page_76thm.jpg
91145 F20101129_AAAVYU halverson_k_Page_95.jpg
23520 F20101129_AAAVTX halverson_k_Page_14.QC.jpg
24517 F20101129_AAAWKQ halverson_k_Page_77.QC.jpg
21056 F20101129_AAAWFT halverson_k_Page_62.QC.jpg
77332 F20101129_AAAVYV halverson_k_Page_96.jpg
82341 F20101129_AAAWAV halverson_k_Page_66.jp2
F20101129_AAAVTY halverson_k_Page_23.tif
6623 F20101129_AAAWKR halverson_k_Page_77thm.jpg
43701 F20101129_AAAVYW halverson_k_Page_97.jpg
94018 F20101129_AAAWAW halverson_k_Page_67.jp2
F20101129_AAAVTZ halverson_k_Page_11.tif
18243 F20101129_AAAWFU halverson_k_Page_26.QC.jpg
22983 F20101129_AAAWKS halverson_k_Page_80.QC.jpg
23367 F20101129_AAAVYX halverson_k_Page_01.jp2
19377 F20101129_AAAWAX halverson_k_Page_68.jp2
22568 F20101129_AAAWFV halverson_k_Page_43.QC.jpg
22485 F20101129_AAAWKT halverson_k_Page_81.QC.jpg
16868 F20101129_AAAVYY halverson_k_Page_02.jp2
83785 F20101129_AAAWAY halverson_k_Page_69.jp2
3322 F20101129_AAAWFW halverson_k_Page_57thm.jpg
58773 F20101129_AAAVWA halverson_k_Page_10.jpg
6355 F20101129_AAAWKU halverson_k_Page_81thm.jpg
11148 F20101129_AAAVYZ halverson_k_Page_03.jp2
107849 F20101129_AAAWAZ halverson_k_Page_70.jp2
23890 F20101129_AAAWFX halverson_k_Page_83.QC.jpg
21159 F20101129_AAAVWB halverson_k_Page_11.jpg
23630 F20101129_AAAWKV halverson_k_Page_82.QC.jpg
6554 F20101129_AAAWFY halverson_k_Page_83thm.jpg
62688 F20101129_AAAVWC halverson_k_Page_12.jpg
6530 F20101129_AAAWKW halverson_k_Page_85thm.jpg
F20101129_AAAWDA halverson_k_Page_35.tif
6453 F20101129_AAAWFZ halverson_k_Page_33thm.jpg
70648 F20101129_AAAVWD halverson_k_Page_13.jpg
24910 F20101129_AAAWKX halverson_k_Page_89.QC.jpg
F20101129_AAAWDB halverson_k_Page_36.tif
73294 F20101129_AAAVWE halverson_k_Page_15.jpg
6528 F20101129_AAAWKY halverson_k_Page_89thm.jpg
F20101129_AAAWDC halverson_k_Page_37.tif
22457 F20101129_AAAVWF halverson_k_Page_16.jpg
25184 F20101129_AAAWKZ halverson_k_Page_90.QC.jpg
23159 F20101129_AAAWIA halverson_k_Page_70.QC.jpg
F20101129_AAAWDD halverson_k_Page_38.tif
75541 F20101129_AAAVWG halverson_k_Page_18.jpg
5702 F20101129_AAAWIB halverson_k_Page_26thm.jpg
F20101129_AAAWDE halverson_k_Page_39.tif
73639 F20101129_AAAVWH halverson_k_Page_20.jpg
6578 F20101129_AAAWIC halverson_k_Page_40thm.jpg
F20101129_AAAWDF halverson_k_Page_40.tif
71138 F20101129_AAAVWI halverson_k_Page_21.jpg
24514 F20101129_AAAWID halverson_k_Page_18.QC.jpg
F20101129_AAAWDG halverson_k_Page_41.tif
68901 F20101129_AAAVWJ halverson_k_Page_23.jpg
6715 F20101129_AAAWIE halverson_k_Page_34thm.jpg
F20101129_AAAWDH halverson_k_Page_42.tif
52509 F20101129_AAAVWK halverson_k_Page_25.jpg




Copyright 2005 by Kent C. Halverson The views expressed in this dissertation are those of the author and do not reflect the official policy or position of the Air Force, Department of Defense, or the U.S. Government


This dissertation is dedicated to my four charismatic leaders-in-training. May they all enjoy a lifelong thirst for learning and love of life.


ACKNOWLEDGMENTS I thank the United States Air Force for giving me the opportunity and financial support to pursue this level of educationI am truly fortunate for having led such a unique and fortunate Air Force career path. I am indebted to Dr. Amir Erez for his crucial guidance and patience in dealing with my long distance dissertation efforts. His keen academic instincts navigated me through the uncertainties of dissertation research, pulled me out of numerous academic death spirals, and gave me the necessary support to complete this dissertation. I thank Dr. Henry Tosi and Dr. Timothy Judge for their advice and insightful comments during the development stage that provided initial direction in my research. Were it not for my firm foundation in multi-level research methods provided by Dr. James Algina, I would never have attempted this dissertation topic. I am also thankful for the technical support and patience of Richard Hughes, who was responsible for translating my survey instruments into a functional Internet-based survey. The numerous last-minute changes made it possible to tap into such a rich source of data. Throughout the three years in school, when I struggled to keep up with the academic workload, my wife adeptly managed all aspects of our household while raising our three, and later four, children. She always seemed to be able to recognize everyones limits, and then provide the balance we all needed. Ironically, her natural managerial skills far exceed anything I could ever hope to achieve, regardless of my newly acquired academic status in the management discipline. I love her and all that she does for me. I cant forget to thank my children, all of whom, whether they knew it or not, sacrificed iv


their time with Dad. Staying late to read stacks of articles was extremely difficult knowing that I had four children waiting to shower me with hugs and kisses each time I step through the door. Finally, and most importantly, I thank God, since without Him none of what I accomplished would have been possible. I hope that I can use my knowledge and skills in a way that will honor Him. v


TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT.........................................................................................................................x CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................6 Hierarchical Causal Model.........................................................................................14 Mechanisms in the Charisma-Satisfaction Link.........................................................15 The Value-Percept Model....................................................................................19 The Job Characteristics Model............................................................................23 Mechanisms in the Charisma-Charisma Link.............................................................27 Charisma Required as an Intermediate Outcome................................................27 Affective Role Modeling.....................................................................................28 Identification........................................................................................................29 Categorization and the Halo Effect.....................................................................31 Mechanisms in the Satisfaction-Satisfaction Link.....................................................35 Happiness and Job Satisfaction: A Recursive Relationship................................36 Emotional Contagion...........................................................................................37 Social Information Processing.............................................................................42 Summary of Research Proposal and Mediation Hypotheses...............................44 3 STUDY 1....................................................................................................................47 Method........................................................................................................................47 Participants..........................................................................................................47 Procedure.............................................................................................................49 Measures..............................................................................................................49 Analysis......................................................................................................................50 Results.........................................................................................................................52 Model 1................................................................................................................53 vi


Models 2a and 2b.................................................................................................54 Models 3a and 3b.................................................................................................55 4 STUDY 2....................................................................................................................58 Method........................................................................................................................58 Participants..........................................................................................................58 Procedure.............................................................................................................61 Measures..............................................................................................................61 Results.........................................................................................................................62 Model 1................................................................................................................62 Models 2a and 2b.................................................................................................63 Models 3a and 3b.................................................................................................63 5 DISCUSSION.............................................................................................................65 Summary of Results....................................................................................................65 A New Approach to Charismatic Leadership.............................................................66 Practical Applications.................................................................................................69 Limitations..................................................................................................................71 Future Research..........................................................................................................73 Conclusion..................................................................................................................75 LIST OF REFERENCES...................................................................................................77 BIOGRAPHICAL SKETCH.............................................................................................86 vii


LIST OF TABLES Table page 3-1 Hierarchical linear model summary.........................................................................53 3-2 Model 1: three-level model of the influence of level-3 leader charisma on level-1 follower satisfaction, Study 1...................................................................................54 3-3 Two-level models of the influence of level-3 leader charisma on level-2 intermediate leader satisfaction (model 2a) and charisma (model 2b), Study 1......55 3-4 Three-level models of the influence of level-2 intermediate leader satisfaction (model 3a) and charisma (model 3b) on level-1 follower satisfaction, Study 1......56 4-1 Model 1: three level model of the influence of level-3 leader charisma on level-1 follower satisfaction, Study2....................................................................................62 4-2 Two-level models of the influence of level-3 leader charisma on level-2 intermediate leader satisfaction (model 2a) and charisma (model 2b), Study 2......63 4-3 Three-level models of the influence of level-2 intermediate leader satisfaction (model 3a) and charisma (model 3b) on level-1 follower satisfaction, Study 2......64 viii


LIST OF FIGURES Figure page 2-1 Multilevel mediation models....................................................................................15 2-2 Leader and followers satisfaction-satisfaction link..................................................37 3-1 Typical fire department organizational structure.....................................................48 4-1 Typical cadet flight structure....................................................................................60 ix


Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MULTILEVEL EFFECTS OF LEADER CHARISMA ON FOLLOWER SATISFACTION By Kent C. Halverson August 2005 Chair: Amir Erez Major Department: Management Leadership literature has established a relationship between leader charisma and follower satisfaction. However, much of this literature hinges on the assumption that charisma is transferred through direct leader-follower relationships. This common, albeit important, perspective has generated valuable empirical research supporting such dyadic theories, but as organizations increase in size it may be unrealistic for leaders to rely on distal dyadic relationships. While distal leaders may have the ability to present poignant speeches that incite followers in the short term, it may be unreasonable to expect such infrequent communications to deeply penetrate follower attitudes and sustain lasting follower satisfaction over the long term. This dissertation suggests that other indirect mediation processes exist that can transfer leader charisma down through hierarchical levels, resulting in a cascading effect of charisma. Specifically, two mediation models are presented that suggest intermediate-level leader satisfaction and charisma mediate the relationship between top-level leader charisma and bottom-level follower satisfaction. x


Using hierarchical linear models in two separate samples, fire fighters and Air Force cadets, the results overall provide support for the charisma-mediating model, but not the satisfaction-mediating model. Such results suggest that future research must not only distinguish direct from indirect transfer of charisma, but also determine which constructs mediate the transfer of charisma. xi


CHAPTER 1 INTRODUCTION In recent years, a large body of research has accumulated in favor of charismatic leader theories. For example, several meta-analyses (Lowe, Kroeck, & Sivasubramaniam, 1996; Fuller, Patterson, Hester, & Stringer, 1996) suggest that charismatic leadership behaviors are related to subjective (r=.73) and objective (r=.30) measures of leadership effectiveness and that these relationships generalize across types of organizations. Although performance-based outcomes dominate the charismatic leadership literature, researchers have also showed that charismatic leaders influence other follower outcomes, such as satisfaction (Dubinsky, Yammarino, Jolson, & Spangler, 1995; Niehoff, Enz, & Grover, 1990, Pillai, Schriesheim, & Williams, 1999; Podsakoff, MacKenzie, Moorman, & Fetter, 1990; Ross & Offermann, 1997). Indeed, in their recent meta-analysis, Judge and Piccolo (2004) report an impressive effect size (r=.58) of the relationship between leader charisma and follower satisfaction. This correlation represents a compilation of results of studies sampling from business sector populations (e.g., Dubinsky et al., 1995; Niehoff et al., 1990; Podsakoff et al., 1990) across multiple and diverse industries (Podsakoff, MacKenzie, & Bommer, 1996), college students (Pillai et al., 1999; Kirkpatrick & Locke, 1996) and public agencies, such as the military (Ross & Offermann, 1997; Percy, 1997). Moreover, these studies show that the relationship between charisma and job satisfaction not only exists at the upper echelons of management, such as CEOs (Niehoff et al., 1990), but at all levels 1


2 (Podsakoff et. al., 1996), including self-directed work teams at the lowest echelon (Butler, Cantrell, & Flick, 1999). Researchers have uncovered some reasons for the influence of leaders charisma on followers satisfaction. For example, researchers have been quite successful in relating distal variables such as leader vision and communication style to predict follower satisfaction (Howell & Frost, 1989; Kirkpatrick & Locke, 1996). For instance, Kirkpatrick and Locke (1996) argued that a leaders vision induces followers to model the leader values and goals, and by this modeling they achieve congruency between their own and the leaders and the organizations values and goals. In turn, these authors argue, the congruency of values between the person and the organization would positively affect follower satisfaction. Indeed in their study Kirkpatrick and Locke (1996) showed that leader vision was positively related to follower task satisfaction. These authors also hypothesized that leaders new ways of approaching work should increase follower intellectual stimulation (Bass, 1985) and therefore uplift followers and increase follower arousal resulting in satisfaction. Here again, their results supported the link between leader intellectual stimulation and follower task satisfaction. A similar argument has been presented by Niehoff et al. (1990) who maintained that charismatic leaders actions make employees tasks more interesting and challenging, and therefore, have a positive effect on employees satisfaction. However, distal variables such as leader vision and communication style only capture part of the story. They tell us very little about the more proximal processes that may explain the influence of leaders on followers satisfaction. For example, it could be argued that general attitudes such as job satisfaction are mainly a combination of daily


3 specific evaluations (see Eagly & Chaiken, 1993; Weiss & Cropanzano, 1996) and moods at work (Judge & Ilies, 2004; Weiss, Nicholas, & Daus, 1999). But it is not clear from leadership theories how leaders influence these daily evaluations that may be so critical to the maintenance of followers satisfaction. Examples of specific evaluations that may influence employee satisfaction are events that occurred during the hiring or promotion process, events associated with the socialization process, an employees daily interaction with peers, and how an employee is treated by his/her supervisor. All these and other events on the job should have a strong effect on our job enjoyment, and therefore our job satisfaction. Similarly, the contagious effect of peers collective moods (Totterdell, Kellett, Teuchmann, & Briner, 1998), the mood of a focal team member (Barsade, 2002), as well as the displayed and experienced emotions of a direct interaction with a leader (Erez, Misangyi, Johnson, LePine, & Halverson, 2005; Lewis & Haviland-Jones, 2000) can also influence our mood at work, and thus, our satisfaction. In contrast, it is unlikely that only one or several critical incidents in which the leader articulates a vision could have a profound effect on individuals satisfaction. It is much more likely that the leader vision and style are somehow translated into more specific influences on followers daily events evaluations and moods at work, and as such, influence followers satisfaction. But leadership theories fail to deal with translation and transmission issues. In other words, while current theories may explain quite well how leaders communication style and vision influence followers satisfaction with specific task or at specific times they do not adequately explain the maintenance of satisfaction over time. Indeed, in their summary of the charismatic leadership literature, House and Aditya concluded that charismatic


4 theories offer inadequate or untested explanations of the process by which the theoretical leader behaviors are linked to, and influencefollowers (1997: 442). Another problem with existing explanations for charisma effects is that these theories usually account for direct and dyadic relationship between a leader and a follower or groups of followers. However, most of the influences of leaders on followers occur through others, and not in a direct one-to-one relationship. Accordingly, we know that if a leader has a constant contact with followers, the followers level of satisfaction will increase because of the leaders vision and communication style (Howel & Frost, 1989; Kirkpatrick & Locke, 1996). But why would a leader that is seen by followers only once a month or even once a year influence followers? Can we really assume that the infrequent communication of vision or the communication style of this leader is not only so poignant and penetrating but also enduring that it can provide and sustain strong maintenance effects on followers without the leaders presence? In this dissertation I will argue that such a strong assumption is not necessary, nor is it necessary for the charismatic leaders to be in constant contact with followers to have an effect on their satisfaction level. That is, because charisma cascades down through the ranks to influence followers and as such, it does not rely solely on singular effects of leaders on followers. Instead, the leader charisma, up the hierarchy, affects bottom-level followers through the intermediate-level leaders direct and constant daily contact with followers. The purpose of this study, then, is to examine the hypothesis that charisma travels down through an organization to influence followers. In order to address this question I will propose a multilevel model describing the charisma transfer process in which intermediary leaders act as charisma conduits. Specifically, this dissertation will


5 focus on the relationship between upper-level leader charisma and follower satisfaction at lower levels as mediated by intermediate-level leader variables. In order to test the propositions of this dissertation I conducted two studies, utilizing two different kinds of leader-follower relationships in two very different populations. Merging the results from these two studies should help confirm the validity of my propositions.


CHAPTER 2 LITERATURE REVIEW Almost thirty years ago, Locke (1976) counted in his review of the job satisfaction literature an overwhelming number of job satisfaction studies, conservatively estimating an annual average of over one hundred studies involving the construct. Since then, satisfaction continues to be a popular construct in the research literature, with over 10,000 more articles identified since 1985 in the PsychInfo database. The literature on leadership is no doubt just as robust. The charismatic leadership literature has experienced a relatively strong growth spurt in the last decade, with 410 articles published since 1996 (this includes transformation leadership, which some researchers consider to be synonymous with charismatic leadership (Judge & Piccolo, 2004)). With such prolific histories, one would expect these two research streams to cross frequently. However, only a relatively small number of studies focus on the relationship between leader charisma and follower satisfaction. In their meta-analysis, Judge and Piccolo (2004) found only 18 studies (n = 5,279) that measured the effects of leader charisma as an antecedent to follower satisfaction. Nonetheless, these studies revealed a quite impressive relationship, with an estimated true score correlation of .58 (p<.01). Not only does the compilation of these studies reveal an impressive effect size of the relationship between charismatic leadership and job satisfaction, but the studies themselves were based on samples drawn from relatively diverse settings and situations, lending credibility and validity to the external validity of the result. A majority of the studies in this meta-analysis used samples drawn 6


7 from business sector populations (e.g., Podsakoff et al., 1990; Dubinsky et al., 1995; and, Niehoff et al., 1990). Although the main purpose of these studies was not always to directly investigate the relationship between charismatic leadership and satisfaction, they nonetheless revealed robust results across a variety of industries. For example, Podsakoff et al. (1990), in their attempt to identify mediators between charisma and organizational citizenship behavior (OCB), sampled 988 petrochemical company employees and found that leader charisma was an antecedent to follower satisfaction, but that this satisfaction failed to mediate the relationship between charisma and OCB as hypothesized. Dubinsky et al. (1995), in their study of sales managers, also found a correlation between transformational (i.e., charismatic) leadership and follower satisfaction. Moreover, these relationships have been found not only at the dyadic link but also at several levels. For example, Niehoff et al. (1990) found that charisma of top leaders (i.e., CEOs) in the insurance industry explained 36% of the variance in job satisfaction of the employees. Two of the studies were based on samples drawn from college student populations. Pillai et al. (1999), in their attempt to identify the mediating effects of justice and trust on the relationship between charisma and satisfaction, found a statistically significant correlation between leader charisma and follower satisfaction, but failed to reveal the hypothesized mediating effects. Kirkpatrick and Lockes (1996) study involving college students further supported the charisma-satisfaction relationship. These authors manipulated charisma in a laboratory environment and found that different components of charismatic leadership (i.e., vision, intellectual stimulation) were related to participants satisfaction with the task. This study is particularly important because it was the only laboratory study in which the charisma-satisfaction link was tested directly.


8 Two other studies were based on samples drawn from military population. Ross and Offermann (1997), in their efforts to study the effects of transformational leadership on performance, found a statistically significant correlation between leader charisma and follower satisfaction among Air Force Academy cadets and their commissioned officers in charge. Percy (1997) obtained results (r = .54) from a military sample that were consistent with the overall meta-analytic results of Judge and Piccolo (2004). While only a few of the 18 studies identified in Judge and Piccolos (2004) meta-analysis were reviewed here, it is evident that the literature has clearly established the relationship between leader charisma and follower satisfaction. At its inception, the charisma concept (Weber, 1947) was only applied to upper-echelon leaders. However, a few decades ago, organization researchers obtained empirical evidence (Bass, 1985; Shamir, House, & Arthur, 1993) which refuted earlier claims that leader distance was a prerequisite for charisma (Etzioni, 1961; Katz & Kahn, 1978), thus increasing the potential span of application of the charismatic leadership theory. Thus, an important aspect of the charisma-satisfaction link is that it operates at varying leader-follower distances, both social and physical (Antonakis & Atwater, 2002). For instance, the link between charisma and satisfaction has been found not only between close leader-follower dyads such as between a supervisor and his direct subordinates (e.g., Podsakoff et al., 1996), but also between more distant dyads such as between a CEO and lowest level echelon employees (e.g., Niehoff et al., 1990). Consequently, the charisma-satisfaction link has been well established as one that can operate in differing contexts and also across multiple hierarchical levels.


9 However, as charismatic leader theories are applied to larger organizations, it quickly becomes evident that the distal dyadic relationships between top-level leaders and lower-echelon followers are significantly different from the proximal dyadic relationship found between a supervisor and an immediate subordinate. One significant assumption in this well-established relationship between leader charisma and follower satisfaction is that the leader has sufficient contact with followers (in frequency and in intensity) at the dyadic level in order for the charismatic influence process to occur on a relatively regular basis. A relatively strong dyadic relationship between leader and follower creates an available and often convenient path for various (e.g., written or spoken) and diverse (e.g., verbal or nonverbal) forms of communication (Antonakis & Atwater, 2002). Such lines of communication allow the charisma process to operate on followers on a regular basis, for instance via criticism of the status quo and communication of vision (Conger & Kanungo, 1987), or developing follower affection, trust, and loyalty for the leader (House, 1977). What this assumption about direct contact implies then is that outside of the proximal dyads, charismatic leaders are constrained in their ability to communicate with followers and therefore influence them. This ability to influence directly may be especially important for the followers satisfaction, or more specifically to the maintenance of follower satisfaction. Satisfaction is no doubt a dynamic attitude that can improve or diminish over time as affected by numerous events and agents (Locke, 1976). Followers are continuously evaluating their work environment and the agents therein to assess their satisfaction. Although possible for short term durations, it is unreasonable to expect that any one event or agent can provide a sustained impact on satisfaction. Consequently, charismatic leaders in distal


10 positions, while able to affect follower satisfaction in the short term, should be less likely to maintain follower satisfaction in the long term. Meindl (1990) suggests that charismatic effects cannot be sustained by leaders themselves, but by other social process such as intermediary personnel. Accordingly, as organizations become large, the direct influence of upper-level leader charisma on lower-level follower satisfaction should be diminished. That is because, as the span of control for the leader increases, direct contact with subordinates becomes increasingly limited. Top-level leaders are a limited resource in an organization, subjected to time and geographic restraints that prevent them from being in all places at all times. This situation usually prevents leaders from having direct contact and control of all subordinates within the organization. As a result, the leader must delegate responsibility and disseminate information through some organizational lines of communication or chains of command. While diverse forms of communication are available (e.g., policy memorandums, emails), it is unlikely that they can provide the necessary direct and consistent conveyance of charisma required to influence follower satisfaction over the long term. Nonetheless, the extant literature clearly demonstrates that leader charisma influences follower satisfaction across multiple organizational levels. The question is what is the process by which this happens? Recently, there has been some effort in the charismatic literature to differentiate between close and distant leadership (Shamir, 1995; Yammarino, 1994; Antonakis & Atwater, 2002). This research emphasizes the importance of looking beyond dyadic relationships to group-level theories (Shamir, Zakay, Breinin, & Popper, 1998). Considering the potential widespread effects leader charisma can have on large groups of


11 people, many of which might have little or no direct contact with the leader, it is likely that some mediation process exists that transmits the effects of leader charisma down to the lowest echelon. While such underlying influence processes remain unclear (Yukl, 2002), researchers have suggested that charisma cascades down through an organization (Waldman & Yammarino, 1999) or can have a falling domino effect (Bass, Waldman, & Avolio, 1987). Consequently, according to this model of leadership, the formal chains of command act as lifelines for top-level leadership. Leaders of all hierarchical organizations depend on teams of frontline subordinates to delegate tasks down to the next lowest hierarchical level. Without command and control of subordinates who are ultimately responsible for the work performance at the worker level, it would be almost impossible for a leader to pilot the ship, so to speak. These intermediary personnel seem to be critical links in the charismatic leadership process across multiple hierarchical levels. Accordingly, any explanation of the link between upper-level leaders charisma and bottom-level followers satisfaction needs to account for the effects of intermediary personnel. According to this cascading model of charismatic leadership, followers of upper-level leaders become leaders for the next lower echelon, and those followers become leaders for lower-level followers in a downward manner to the bottom level. It would seem logical then that the outcomes of the influence of upper-level leaders on the intermediate-level population would be good predictors of variables of interest at the lower-level population. Unfortunately, none of the existing charisma theories are able to accurately model multilevel cascading behavior of charisma. For example, researchers have proposed several dyadic charisma process theories (Shamir, et al., 1993; Gardner &


12 Avolio, 1998) in which one set of leader behaviors affects a different set of follower behaviors as moderated by follower self-concept or impression management, respectively. This theory then explains that the process operates in the dyadic relationship but does not explain how charisma will influence followers beyond this relationship. Jacobsen and House (2001) present a complex system dynamics model to simulate the process of growth and decay of charisma over time. However, the systems dynamics approach fails to identify specific relationships or transmission processes, and instead provides a heuristic simulation model that generates relative differences in behavior among combinations of input variable intensities. Lord and Emrich (2001) also present propositions dealing with charismatic leadership process. Although their study focuses on a cognitive dimension of charismatic leadership, the authors highlight the fact that most of the charisma studies neglect intermediary processes (Lord & Emrich, 2001) and instead focus on outcomes (e.g., DeGroot, Kiker, & Cross, 2000). In an attempt to integrate the literature on leadership distance, Antonakis and Atwater (2002) propose a model in which subordinate level leaders mediate leaders in distal positions. Unfortunately, their discussion on this important cascading mediation process was limited to one paragraph in which little is proposed. Finally, Meindl (1990) and Madsen and Snow (1983) suggest that charisma is mediated by informal secondary leaders, possibly even resulting in the exclusion of the leader all together. Thus, while most agree that the affects of charismatic leadership transcend hierarchical levels, very little empirical research exists on the process by which charisma travels down through an organization.


13 In my dissertation I will attempt to correct for this situation by developing a series of hypothesized links that will explain the connection between upper-level leader charisma and lower-level follower satisfaction. However, before I develop these hypotheses a note on my method of investigation is required. In order to study the propositions of my study in which group-level variables influencing individual-level outcomes, a multilevel perspective is required (Bliese, Halverson, & Schriesheim, 2002). The multilevel aspect of leadership is a methodological issue that complicates the quest for clarifying the charismatic leadership process, but at the same time it is recognized by many researchers that adequate leadership theories cannot ignore the multilevel perspective inherent in the leadership-followers relationship. In fact, The Leadership Quarterly journal published a special issue in 2002 specifically devoted to the topic of multilevel leadership research. Thus, researchers have recently begun to use multilevel methods to study the multilevel effects of leadership (Dvir & Shamir, 2003; Bliese et. al., 2002; Schriesheim, Castro, Zhou, & Yammarino, 2001). Notwithstanding that, these studies only attempt to discriminate the dyadic effects from the group level effects (Avolio & Yammarino, 1990). While this is an important research pursuit, to a large extent, and in line with the arguments I presented previously, the multilevel charisma process could not be just contained within two levels of analysis. However, beyond the dyadic perspective, to the best of my knowledge, only one study has analyzed a three-level model in an empirical leadership study. Dvir and Shamir (2003) studied the effects of followers in influencing leader behavior. This study utilized the three level data set to discriminate the affects of close (i.e., level-2) followers versus distant (i.e., level-3)


14 followers. Results indicated that follower dispositions have a role in leadership perception, but the study was void of discussion of multilevel charismatic processes. Hierarchical Causal Model A hierarchical causal model was hypothesized that includes relationships among upper-level leader charisma, intermediate-level leader charisma, and satisfaction, and lower-level follower satisfaction. Figure 1 depicts the links contained in the hypothesized multilevel model. The links included in this model are 1) the relationship between leader charisma-follower satisfaction, proposed between top-level leader to intermediate-level leaders and in turn a charisma-satisfaction link from intermediate-level leader to bottom-level level followers, 2) the relationship between leader charisma and intermediate-level leader charisma, and 3) the relationship between intermediate-level leader satisfaction and bottom-level follower satisfaction. In the following section, I will propose and discuss the theories that provide the necessary foundation for each of the suggested multilevel links. Finally, these links will be connected to form two suggested paths as shown in Figure 2-1, along which upper-level leader charisma travels as it cascades through the ranks to influence bottom-level followers satisfaction.


15 Figure 2-1. Multilevel mediation models Mechanisms in the Charisma-Satisfaction Link As previously discussed, substantial empirical evidence exists to support the relationship between leader charisma and follower satisfaction, as observed in diverse settings, different populations, and different experimental designs (Judge & Piccolo, 2004). Furthermore, this relationship has been observed in proximal dyadic leader-follower relationships (Podsakoff et al., 1996) as well as in distal dyads (Niehoff et al., 1990). While some process-explanations exist about the charisma-satisfaction link in direct leader-followers relationships (see Kirkpatrick & Locke, 1996), none exist for this link between distant leaders and followers. Thus, the purpose of this section is to theorize about the potential mechanisms that could explain the relationship between upper-level leader charisma and lower-level follower satisfaction. In this section I will first discuss


16 why the difference between distal and proximal leaders effects on followers job satisfaction is important. I will then utilize two well-established theories of satisfaction, the value-percept model and the job characteristics model to explain how indirect charismatic leaders influence follower satisfaction. Most researchers would agree that the behavior displayed and the actions taken by charismatic leaders influence the affective state of followers (House, 1977: Bass, 1985). Leaders, and especially charismatic leaders, create critical events that influence followers beliefs, attitudes, and moods. However, the effects of leader charisma, if only displayed very infrequently, may wane over time as individual attitudes and affective states return to their normal steady-state. Research has shown that individuals maintain relatively stable affective attitudes (Tellegen, 1985; Watson & Clark, 1984; Watson & Slack, 1993). This research suggests that individuals are affectively predisposed, influencing their positive affective attitudes (e.g., optimism) as well as negative affective attitudes (e.g., pessimism). The research on job satisfaction shows similar dispositional dependence (Judge & Larsen, 2001). Although individuals attitudes can be temporarily influenced by daily events (Locke, 1976; Weiss & Cropanzano, 1996), these attitudes eventually return to their steady-state level (Judge & Ilies, 2004, Erez, Mitchell, Jackson, Fanelli, & Judge, 2005). For example, Staw and his colleagues found that job satisfaction in individuals is relatively consistent over time (Staw, Bell, & Clausen, 1986), and across contexts (Staw & Ross, 1985). This consistency suggests a steady-state level of satisfaction. However, Staw et al. (1986) also caution that their results do not suggest that individual satisfaction levels do not vary or are not vulnerable to contextual factors. Indeed, job satisfaction theories propose that satisfaction is affected by agents and events


17 (Locke, 1976; Weiss & Cropanzano, 1996). For example, empirical data indicate that events that induce positive affective states can positively influence satisfaction (Kraiger, Billings, & Isen, 1989; Watson & Slack, 1993; Brief, Butcher, & Roberson, 1995). However, other findings also suggest that, longitudinally, changing circumstances have less of an effect on satisfaction and that in the absence of any affect-increasing events, satisfaction stays relatively stable (Staw & Ross, 1985). In other words, these findings may suggest that events should have an effect on job satisfaction but once the event manipulation is removed or fades away, the individuals satisfaction should return to its normal, steady state. Thus, graphical representations of longitudinal attitude levels may resemble a sinusoidal pattern anchored at the steady-state level, where the frequency and amplitudes of peaks and valleys correlate with the timing and intensities of contextual influences, such as exposure to charismatic leadership. Accordingly, individuals can remain at a relatively stable affective state, but events can temporarily influence these semi-permanent affective states. All this means that if charismatic leadership is responsible for changing follower attitudes, it seems as though these attitudes would require regular maintenance to sustain the attitudes over the long term. Based on the propensity of satisfaction to be stable, it seems that distant leaders would be unlikely to attain much of a lasting effect on followers satisfaction. Although there may be unique instances where a distant charismatic leader gives a speech that changes someones life, in daily organizational life it would not be likely to be a common event and could even be perceived as an anomaly. This, of course, does not mean that upper-level leaders will never influence followers directly and that their only influence will be through others. However, it seems that at least part of their influence on followers


18 will be through intermediate personnel and, therefore, this link needs to be investigated. In contrast, proximal charismatic leaders, by way of daily meetings and other communications, have the ability to use their charismatic influence more frequently. For example, charismatic college basketball coaches are often highlighted for their effects on team performance. They are in a position to provide more frequent charismatic influence throughout each practice or game. However, even they are limited in the amount of influence they can exert on their athletes off the court. On-court success does not necessarily lead to off-court performance, perhaps because the athletes lack required leadership off the court. Only the most respected coaches, such as Mike Krzyzewski from Duke University, seem to produce players who are successful both on and off the court. Such charismatic coaches seem to be successful in developing student athletes who seem to be driven by more intrinsic rewards than tallys in the win column, and the effects are enduring. It is unlikely that such results can be achieved only by influence on the courtit most likely requires influence on a more frequent basis. Thus, proximal charismatic leaders may incite through vision and subsequently influence followers on a daily basis. Proximal charismatic leaders then can attempt to align follower values and goals with their vision (Conger, 1989). Although the above discussion describes the ease of influence of proximal leaders versus distal leaders to influence followers satisfaction it does not describe the process by which followers are influenced by proximal leaders. The following two sections discuss in detail two of these processes that could help establish a link between charisma and satisfaction.


19 The Value-Percept Model Discrepancy theories that deal with the congruity between a person and an organization can be readily utilized to explain charismatic leadership influences on followers satisfaction. For example, supplemental fit, one of the main dimensions of P-O fit (Kristof, 1996), occurs when individuals adopt or possess values of other individuals in the organization (Muchinsky & Monahan, 1987). When the values of an individual are closely aligned with those of the organization, the person experiences a high level of P-O fit, and empirical research indicates that P-O fit predicts job satisfaction (Chatman, 1991; OReilly, Chatman, & Caldwell, 1991). Thus, one of the main ways by which charismatic leaders can influence followers satisfaction is by changing followers values and aligning them with organizational values. Furthermore, the values of individuals within an organization are combined to form the organizational culture (Schein, 1990), with more congruent values leading to stronger culture (Chatman, 1989). In this way charismatic leaders could also influence the organizational culture as a whole. Both distal and proximal charismatic leaders can reduce the discrepancy between peoples values and the organizational values; however, proximal leaders are also in the unique position to close other gaps that may influence individuals satisfaction. The value-percept model suggests that satisfaction is a function of the discrepancy between ones values (i.e., wants, assuming they are congruent with ones needs) and the value provided by the job, multiplied by an importance factor (Locke, 1976). In other words, individuals evaluate each dimension of their job by comparing what they want to what they are actually getting, and then they give weight to each discrepancy by determining how important each dimension is. Research indicates an inverse relationship between value discrepancy and satisfactionthe smaller the discrepancy between individual


20 values and what values the job provides, the greater the satisfaction (Locke, 1976). For example, window offices tend to be prized fringe benefits in the white collar world. Lets assume for the sake of argument that all employees want a window office. However, in most buildings there is a limited supply of window offices, leaving many with no outside view. Those employees who are assigned a window office have little or no discrepancy between what they want and what they have and should be satisfied. For all of the unfortunate employees who do not get a window office, there is a large discrepancy between what they want and what they are getting. Of these unfortunate employees, those that place a lot of importance on a window office will tend to be the least satisfied due to the great weight they put on this discrepancy. In contrast, employees who assign little importance to a window office may still be satisfied regardless of the discrepancy. Evaluating overall satisfaction requires the accumulation of all the different dimensions and their respective importance factors. The value-percept theory of satisfaction can be readily applied to the charismatic leadership process. First, as mentioned before, charismatic leaders can align followers values with organizational values and therefore they may perceive less discrepancy between what they have and what they want. Indeed, empirical data indicate that leader charisma induces followers to model after the charismatic leader, resulting in the alignment of follower values and goals with the organizational values and goals (Kirkpatrick & Locke, 1996). (One assumption used for ease of communication is that the charismatic leaders values and goals are congruent with the organizational values and goals.) Consequently, follower values that are aligned with the organization values reduce the discrepancy and increase satisfaction (Locke, 1976; OReilly et al., 1991). In


21 fact, some researchers even suggest that a charismatic leaders first task is to identify a discrepancy between the status quo and a desired end state, and then emphasize the importance of that desired end state (Conger & Kanungo, 1987). Charismatic leaders, then, have the ability to persuade followers to take on the organization goals and therefore drastically reduce the gap between what followers want and what they are getting, or could get. As previously mentioned, this reduced gap increases follower satisfaction. Second, one of the main features of charismatic leaders is that they make followers look beyond their narrow interest. In fact, charisma is almost by definition a leaders ability to motivate followers to work for transcendental goals that go beyond immediate self-interests (House, 1977; Bass, 1985). Accordingly, charismatic leaders can not only reduce the discrepancy between followers haves and wants but also can change the importance weight assigned to the discrepancy by followers. If we take the office window example then, the charismatic leader is not very likely to convince a person who wants a window office that he or she actually has it (unless this person is delusional); however, the charismatic leader may be able to convince this person that in the great scheme of things having a window office is not that important. In that way, even if discrepancy still exists between what individuals have and what they want, followers of charismatic leaders may be more satisfied than followers of non-charismatic leaders because they should weight less the importance of this discrepancy. As suggested previously, distant charismatic leaders may be able to only influence followers in the short term. As such, sporadic and limited contact with distant charismatic leaders may result in temporary alignment of followers haves and wants. If leader


22 charisma motivates followers to go above and beyond their self-interest, as research suggests (Bass, 1985), the leader may also temporarily change the importance weight assigned to the discrepancy. However, this change should not last long and therefore should only have a momentary effect on followers satisfaction. As followers adopt new values aligned with the leaders values, it will most likely come at the expense of some other follower values that are eclipsed by the charismatic leaders vision. How long can followers sustain the alignment of their values without sustained charismatic influence? Furthermore, how long will followers tolerate the charismatic leaders vision overshadowing the values that they previously considered important? Since distal charismatic leadership may only provide short-term follower satisfaction, it is unrealistic to expect that followers will sustain long-term sacrificial effort toward the organizational vision. Charismatic leaders involved in proximal leader-follower dyads, which usually allow followers more frequent and consistent contact with the charismatic leader, are in much better positions (Antonakis & Atwater, 2002) to provide the necessary gap reduction and weight change of the value-percept discrepancy among followers. Proximal charismatic leaders, such as those who work in the same office or shop as the followers, have opportunities throughout the day to provide vision, support and motivation for the followers. These charismatic leaders act as agents who can influence values, vision, and motivation. This includes both the alignment of values as well as a change in the importance of certain values. As followers are directed toward and adopt the charismatic leaders values and vision, the discrepancy between the followers values and the charismatic leaders values decreases. As discussed previously, the reduced gap in value


23 discrepancy results in increased satisfaction (Locke, 1976). Furthermore, sustained influence by the proximal charismatic leader should encourage sustained performance by the followers that is above and beyond self-interest (Bass, 1985). When followers are asked to make personal sacrifices in pursuit of the charismatic leaders vision, any increased effort by the followers may last only as long as the charismatic leader promotes his or her vision. As long as the charismatic leader has the ability to influence followers, increased effort on the part of the follower might be sustained. However, without charismatic influence and promotion of the vision, followers will most likely return to their normal behaviors and attitudes (Komaki, Berwick, & Scott, 1978). Proximal leaders then, because they have more direct contact with followers, may be in a better position likely than distal leaders to promote their vision that aligns the followers values with their own values. This proximal influence, of course, can happen at any level of the organization and all it requires is just a direct and frequent contact between the leader and the followers. Thus, upper-level leaders may have proximal relationships with intermediate-level leaders and those in turn may have proximal relationships with the next lower level of followers. As such, the discrepancy reduction of the value-percept model influenced by leader vision can cascade down the organization ranks. The Job Characteristics Model One of the most important influences on job satisfaction is how individuals perceive the work itself. Hackman and Oldman proposed (1975) and tested (1976) a model that suggested that five job dimensions influence personal outcomes such as satisfaction. A meta-analysis of 200 studies indicates a correlation of .63 between the job characteristics model and job satisfaction (Fried, Y. & Ferris, G. R., 1987). The job characteristics model proposes that the five dimensions (skill variety, task identity, task


24 significance, autonomy, and feedback) are mediated by three critical psychological states: experienced meaningfulness of the work, experienced responsibility for outcomes of the work, and knowledge of the actual results of the work activities (Hackman & Oldman, 1975). Three of the job dimensions, skill variety, task identity, and task significance, are mediated by the meaningfulness of the work, while the fourth job dimension, job autonomy, and fifth job dimension, feedback, are mediated by experienced responsibility and knowledge of results, respectively (Hackman & Oldman, 1975). Inherent in most leaders jobs is the ability to control how the tasks are performed (e.g., work breakdown schedules) and who performs them (e.g., hiring, firing, retraining, promotion). Each of the characteristics of a job must be communicated by the leader, all of which could be influenced by the proximal leader, but are not very likely to be influenced by a distal leader. For instance, if a proximal leader notices an employees production decreasing, and finds out that the employee is bored by his or her job, the leader can increase the scope of the task or add new tasks, thus requiring the use of different skill sets and enhancing employee satisfaction. In contrast, a distant leader is not likely to be in a position to notice this decrease in production by a specific employee. Three of the job characteristics, skill variety, task identity, and autonomy, are particularly influenced by proximal leaders. Distal leaders are far removed from intricacies of work at the lowest echelons. Yearly plant visits may be as close as some leaders come to the work that takes place by such workers. Consequently, distal leaders are not familiar with the skill variety associated with all the various jobs, the extent to which a task results in an identifiable piece of work, or the impact of the work relative to the others in close proximity. As a result, distal leaders are poorly positioned to


25 manipulate these three job characteristics to enhance individual satisfaction. In contrast, proximal leaders often have direct responsibility over skill variety, task identity, and autonomy, and are perfectly positioned to manipulate any of these dimensions to enhance individual satisfaction. Furthermore, not only are proximal leaders able to influence, they are also able to provide influence on a consistent basis necessary to sustain individual attitudes like satisfaction. These three dimensions can be influenced by all leaders and not only by charismatic leaders; however, in some instances, proximal charismatic leaders may be more adept than non-charismatic leaders in changing employees evaluations. For example, if the charismatic supervisor senses any dissatisfaction among these three dimensions for an individual, he or she may be able to manipulate and modify the characteristics of the job to make it more intellectually stimulating (Bass, 1985) in order to reduce or mitigate the dissatisfaction. In addition, although, skill variety, task identity, and autonomy are usually based on established job designs, in rare occasions in organizations which are experiencing a crisis, charismatic leaders may also have discretion to make radical changes (Conger & Kanungo, 1987) to job designs, facilitating a better job fit for followers. The two characteristics that seem most vulnerable to leader charisma are task significance and feedback. By definition, charismatic leaders would emphasize and reveal the importance of seemingly menial tasks and how they impact the greater good or overarching vision (House, 1977; Bass 1985). Furthermore, charismatic leaders would provide encouragement and positive feedback in an effort to increase self-efficacy and performance (Bass, 1985). While task importance and feedback are not exclusively


26 dependent on leader proximity and distal charismatic leaders such as CEOs may influence task importance and feedback from time to time, because of their limited ability to communicate directly with followers, it is not likely that such infrequent communication can sustain follower attitudes over long durations. For instance, assembly line workers involved in mundane, repetitive tasks performed on the conveyor belt may be temporarily excited or encouraged by a CEO plant visit. However, it is unlikely that the CEOs visit will have sustained follower attitudes a week later. As research indicates (Komaki et al., 1978), attitudes and behavior eventually return to some normal level. Charismatic leaders in proximal dyadic relationships with followers will be in contact with followers on a regular basis, which will allow these leaders to not only routinely monitor but also modify these two job characteristics as necessary to sustain follower satisfaction. To build on the previous example, a supervisor in that same assembly plant can give daily pep talks to the assembly line workers to influence their perception that the task is significant and provide positive and optimistic feedback to sustain positive attitudes. Even if charismatic leaders are not able to tailor all the job characteristics, the leaders retain the ability to change the relative importance of job characteristics, emphasizing some characteristics over the others. If employees complain about one characteristic of the job, an immediate supervisor can possibly increase one of the other job characteristics, thus reducing or even negating the dissatisfaction with the one characteristic. This type of hands-on leadership cannot be accomplished from afarthe leader must be closely and routinely involved with followers in order to perform such tailoring of follower job characteristics (Antonakis & Atwater, 2002). Consequently,


27 proximal charismatic leaders can influence follower satisfaction over the long term through routine monitoring and modification of the salient job characteristics. According to this theory of satisfaction, without the proximal dyadic leader-follower relationship and frequent flow of communication between the two, leader charisma will not affect follower satisfaction. The two theoretical frameworks of the value-percept model and the job characteristics model discussed above along with empirical evidence as summarized in the meta-analysis by Judge and Piccolo (2004) suggest that H1: Leader charisma will influence follower satisfaction at both the distal and the proximal levels. Mechanisms in the Charisma-Charisma Link Charisma Required as an Intermediate Outcome The previous section argued that leader charisma can influence follower satisfaction both between top-level leaders and intermediate-level leaders, and between intermediate-level leaders and bottom-level followers. This argument suggests the antecedent to bottom-level follower satisfaction is intermediate-level charisma. What is now needed is a theoretical justification for the upper-level leader charismaintermediate-level leader charisma link. This dissertation suggests that top-level leader charisma can influence intermediate-level charisma. This is a critical link on which the cascading charisma concept is based. This next section will explain why intermediate-level leaders who are exposed to top-level leader charisma will become more charismatic themselves. Strong leader-follower dyads enhance charismatic influence due to their close proximity and frequent exposure to both verbal and non-verbal communication displayed


28 by top-level charismatic leaders (Antonakis & Atwater, 2002). In order for this charismatic influence to spread further down through the organizational hierarchy, since followers in more distal organizational positions are less likely to be exposed to or be directly affected by top-level leader charisma, the effects of top-level leader charisma observed at the intermediate level must subsequently initiate a similar charismatic process between intermediate-level leaders and the next lowest echelon of followers (Meindl, 1990). Methodically speaking, the dependent variable in the first step of the cascading process becomes the independent variable for the next cascading step, and so on. In other words, whatever causes top-level leader charisma to rub off on the intermediate-level leaders, a similar process is necessary for the intermediate-level leader to rub off on lower-echelon followers. If top-level leader charisma is proposed to cascade all the way down to the lowest echelons, then the charisma must theoretically transfer to the intermediate-level leaders. While those involved in the earliest origins of charisma assumed that charisma was a divinely inspired gift (Conger & Kanungo, 1987), recent studies show that individuals can be trained to display charisma behavior (Howell & Frost, 1989). This is promising for those individuals who lack natural charismatic tendencies, but who are in positions that require them to act as conduits of charisma. Affective Role Modeling There are several proposed charisma mechanisms that may be able to explain the transfer of charisma from one individual to another: affective role modeling, identification, and social categorization. One of the main components of Houses (1977) charismatic leadership theory is that charisma influences an affective dimensionthe followers like the charismatic leaders. Charismatic leaders tend to be energetic, supportive, and provide optimistic outlooks (Conger & Kanungo, 1987; Bass 1985), and


29 it is reasonable to assume that most individuals develop affection for leaders who display such behaviors. Individuals also aspire to be like those they like, and as a result, this affective process can result in role modeling the behavior they observe (House, 1977). An underlying assumption in charismatic leadership theories is that they rely on the expression and subsequent alignment of emotion, values, and self-concepts between leaders and follower (Connelly, Gaddis, & Helton-Fauth, 2002). As the emotions, values, and self-concepts of followers align with those of their leaders, role modeling can result (Shamir et al., 1993), perhaps making charisma contagious (Cherulnik, Donley, Wiewel, & Miller, 2001). Consequently, followers of charismatic leaders should tend to role model those charismatic characteristics and behaviors and become more charismatic like the leader. Identification The most important aspect of charismatic leadership theories may be the identification process (Conger, 1989; Willner, 1968; House, 1977; Shamir, et al., 1993; Bass, 1988). The identification process is similar to role modeling, but it involves more intimate psychological involvement. While role modeling simply involves followers molding their attitudes and behavior with those of the leader, the identification process involves a deeper psychological bond in which the followers belief about a leader becomes self-defining (Kark, Shamir, & Chen, 2003). Followers of charismatic leaders can project themselves into their leaders situation, and as a result, are likely to experience similar feelings (Bandura, 1969; Stotland, 1969). Specifically, followers of charismatic leaders are perceived to have a strong identification with that leader (Bass, 1988). Subordinates may even idolize their leaders in attempts to become like them (Yukl, 2002). Furthermore, this identification process is suggested as a major mechanism


30 in the charisma process that influences follower motivation (Shamir et al., 1993). Followers who are motivated will most likely display increased levels of energy, enthusiasm, and optimism, all of which are characteristic of charismatic leaders (Conger & Kanungo, 1987). The identification process as applied to charismatic leadership suggests many beneficial follower outcomes such as aligned values, and increased energy and motivation (Bass, 1985). While the individual follower outcomes are meaningful to charismatic research, it is the combination of all the follower outcomes in addition to identification that may make the follower seem more charismatic to others. For instance, once a follower adopts a charismatic leaders behaviors and attitudes, the identification process may cause these desirable outcomes to coalesce within the follower, resulting in follower charisma. As has been the case in much of the previous discussion, leader proximity moderates the identification process (Antonakis & Atwater, 2002), meaning followers in close proximity are likely to have greater exposure to the charismatic leader, thus facilitating the identification process. The discussion to this point has concentrated on the direct effects of charisma, mainly emphasizing how intermediate-level leaders can become more charismatic themselves. Each of these theories explanations then requires the intermediate-level leaders to change their behavior and attitudes to some extent. But what if the intermediate-level leaders do not posses the ability to change their behaviors and attitudes? Does this mean that a leaders charisma cannot influence these emotionally oblivious or impervious intermediate-level leaders? And does it mean that in such instances charisma will not be transferred down? Not necessarily because intermediate


31 level leaders may still be influenced indirectly by processes requiring neither changes in intermediate-level leader behaviors and nor changes in their attitudes. Categorization and the Halo Effect The theory of leadership categorization (Lord, Foti, & Phillips, 1982) suggests that followers rely on implicit leadership theories to categorize leaders, or more specifically, to distinguish leaders from non-leaders. Cognitive researchers suggest that individuals tend to categorize information based on specific features they initially observe (Ashcraft, 1989). Categories in an individuals memory can usually be represented by a prototype (Rosch, 1975), which is a stimulus that includes a majority of the typical features common to most stimuli within that certain category (Ashcraft, 1989). A new stimulus will contain features that are considered either typical or atypical relative to the prototype already stored in a category, resulting in an evaluative process which determines if the new stimulus belongs in that specific category (Ashcraft, 1989). There is not necessarily a finite set of dichotomous criteria that determines if a stimulus belongs in a certain category (Feldman, 1981), and instead, natural categories (Rosch, 1973) are bound by fuzzy borders controlled by certain loosely-defined features and assigned various levels of importance (Ashcraft, 1989). For example, although there is probably a general agreement on the category of dogs based on biological definitions, some individuals may not consider certain species of dogs (e.g., Chihuahua as dogs) in their internal categorization. In this case, these individuals who dont consider Chihuahuas to be dogs, regardless of the well-established canine classification system, may assign a high level of importance to dog features such as size, and therefore, all dogs that do not meet their self-defined fuzzy criteria are not categorized as such.


32 Because the categorization process relies on the central features of a prototype to determine membership of stimuli within a category, the prototype can influence or become representative of other stimuli in the category (Feldman, 1981). Seldom do stimuli stored in memory contain information on all the salient features of the prototype (Feldman, 1981). When a stimulus with incomplete information is recalled from memory, often the recall can automatically include information adopted from the prototype that was never possessed by the original stimulus (Feldman, 1981). Feldman (1981) states, When specific information (about the stimulus) is not available, the prototype is used for guessing. He further suggests that this automated process leads to bias which is functionally identical to stereotyping involving underor overevaluations of a stimulus within a certain category (Feldman, 1981). The categorization process may be applied to followers perception of leaders and those individuals closely associated with the leaders, resulting in unintentional bias and overevaluation of those individuals associated with the leader. Lord et al. (1982) introduced a theory of leadership categorization which suggests that followers implicit leadership theories influence how leaders are cognitively categorized to distinguish leaders from non-leaders. Followers evaluate their leader, and if the leaders traits and behaviors are congruent with their implicit concept of a leader, then the leader will be cognitively categorized as such. Empirical data supports this leadership categorization theory, indicating that individuals rely on prototypciality of leader behaviors to form leader perceptions (Lord, Foti, & De Vader, 1984). In other words, followers perceptions of leaders will be based in part on how the followers cognitively categorize the leaders.


33 The more prototypical traits displayed by the leader, the more likely the leader will be categorized as a leader. It is also possible to extend this concept to multiple leaders at different hierarchical levels. Charismatic leaders are inherently prototypical due to the influence they have on followers. Thus, although other leader prototypes may exist for followers, the most pertinent leader prototype within their respective organizational context is likely to be the top-level charismatic leader. Intermediate-level leaders are naturally required to interact relatively frequently with the top-level leader for obvious reasons, such as to disseminate information from top-level leaders. Followers most likely directly observe or are at least aware of the interactions between intermediate and top-level leaders. Such relationships between intermediate and top-level leaders may be perceived by followers as being close. Consequently, it is likely that intermediate-level leaders will be categorized as a leader in the same group as the prototypical top-level charismatic leader. According to the categorization theory, because intermediate-level leaders are categorized with top-level charismatic leaders, the intermediate-level leaders may be perceived as adopting some of the salient traits and behaviors of the top-level charismatic leader. In other words, if followers categorize intermediate-level leaders in the same category as the prototypical top-level charismatic leaders, followers are likely to assign some of the same charismatic traits of the top-level leader to the intermediate-level leader, resulting in the transfer of charisma. One other influence in the categorization process is affect. Affective states have an impact on the social categorization process, such that individuals in positive affective states tend to categorize nontypical exemplars of the category as belonging to the


34 category (Isen, Niedenthal, & Cantor, 1992). In other words, charismatic leaders create an atmosphere of optimism, enthusiasm and encouragement, which can influence followers moods. In turn, the induced or enhanced positive affective state of the followers influences their cognitive categorization ability, allowing them to categorize their intermediate-level leaders with the top-level charismatic leaders. Consequently, since the followers assign the intermediate-level leaders to a category in which a charismatic leader is the prototype, it is likely that the charismatic characteristics common to the prototype will be recalled by the followers and associated with the intermediate-level leaders. This additional affective influence further supports the possibility that leader charisma influences follower charisma. The categorization process is similar to the halo error effect (Thorndike, 1920), which suggests that general perceptions of an individual influence specific ratings of that individual. For instance, likeable (unlikable) leaders are likely to be rated high (low) on a specific behavior such as providing timely feedback simply because the raters overall impression of the leader is positive (negative). The halo effect may extend beyond the top-level leader down to subordinate intermediate-level leaders. Followers may associate the intermediate-level leaders with the top-level leaders, and therefore, any general perception the followers have about the top-level leaders may influence their perceptions of intermediate-level leaders. Contrary to much of the halo error literature describing the negative consequences (Murphy, Jako, & Anhalt, 1993), halo error in this case may actually be beneficial to a charismatic leader. For instance, followers of top-level charismatic leaders may view the intermediate-level leaders in the same light as the top-level charismatic leaders due to the categorization process and subsequent halo effects.


35 Followers may perceive intermediate-level leaders as charismatic simply because they fall within the halo of the top-level charismatic leader. A key point to emphasize for the categorization and halo processes is that neither process influences intermediate-level leader behavior or attitudes, but instead relies on follower perceptions. Although the charisma construct is a follower-based perception, it is a perception usually based on observed leader behaviors. The halo effect requires no change in intermediate-level leader behavior, yet follower perceptions of charisma may still exist. Consequently, the charisma of the upper-level leader may lead to charisma in the intermediate-level leader, absent any significant charismatic behavior displayed by the intermediate-level leader. Thus, whether intermediate-level leaders are indeed behaving more charismatically when they are in contact with upper-level charismatic leaders or they are just perceived to be more charismatic by association with the upper-level charismatic leader, both paths lead to the same hypothesis: H2: Upper level leader charisma will be positively related to intermediatelevel leader charisma. Mechanisms in the Satisfaction-Satisfaction Link Two mechanisms are proposed to establish the relationship between leader satisfaction and follower satisfaction. The first mechanism relies on a reciprocal relationship between happiness and satisfaction within an individual (Judge & Hulin, 1993; Judge & Locke, 1993) and the contagion of happiness between individuals (Hatfield, Cacioppo, & Rapson, 1994). Second, the satisfaction-satisfaction link could also be explained by the social process described by Salancik and Pfeffer (1978) in their social information theory of job satisfaction.


36 Happiness and Job Satisfaction: A Recursive Relationship Figure 2 represents the process by which leaders satisfaction may influence followers satisfaction. Research has shown that subjective well being (i.e., life satisfaction or happiness; Diener, 1984) and job satisfaction share a recursive relationship in that each affects the other (Judge & Hulin, 1993; Judge & Locke, 1993). However, the causal paths linking the two constructs in each direction are supported by differing hypotheses. The causal link between job satisfaction and happiness is based on the suggestion that work is a very important part of most individuals lives (Judge & Locke, 1993). More specifically, most individuals devote a significant amount of the time in their lives to their work, and as a result, the job becomes an important source of identity for the individual (Judge & Locke, 1993). In other words, individuals who are very satisfied with their jobs will tend to feel good about themselves and be happy. Thus, job satisfaction affects happiness because of a spillover effect from job to life. Accordingly, leaders who are satisfied with their jobs will also tend to be happy in general, justifying the first link represented in Figure 2. The assumed causal relationship between happiness and job satisfaction is mainly supported by an idea rooted in cognitive psychology. Happy individuals, or those who were induced with positive mood, have been found to store and recall information differently than those in neutral moods or from unhappy individuals (see Isen, 2000). In particular, positive affect has been found to cue a wide range of positive material in memory and therefore happy people tend to recall more positive information stored in memory (Isen, Shalker, Clark, & Karp, 1978). Consequently, in the process of evaluating the level of satisfaction with their job, individuals who are happy will very likely recall positive memories and not negative memories, resulting in an increased positive evaluation of their job satisfaction. Indeed,


37 Kraiger et al. (1989) have found that people in positive affect tended to evaluate their jobs as more interesting and were more satisfied with their jobs. Accordingly, happy followers should be more satisfied with their jobs as is suggested by the third link in the model showed in Figure 2-2. Figure 2-2. Leader and followers satisfaction-satisfaction link While this recursive relationship between happiness and job satisfaction within an individual explains part of the satisfaction-satisfaction link these recursive relationships are intra-individual processes that by themselves cannot explain the transfer of satisfaction from leaders to followers. What is necessary is a process that external to the individual that can explain how satisfaction or happiness, since either should influence the other, can transfer between individuals. The emotional contagion mechanism, represented in the second link of Figure 2, may be able to explain how satisfaction can transfer between individuals. Emotional Contagion A growing body of literature suggests that emotions are contagious (Hatfield et al., 1994). Emotional contagion is a multilevel and multidimensional mechanism that is psychophysiological, behavioral, and social (Hatfield et al., 1994). Hatfield et al. (1994)


38 describe several mechanisms that can transfer emotions among individuals; however, in this dissertation I will only describe the most relevant process of transferring happiness between proximal leader and followers the mimicry/feedback mechanism. This mechanism was defined by Hatfield et al. (1994) as the tendency to automatically mimic and synchronize facial expressions, vocalizations, postures, and movements with those of another person and, consequently, to converge emotionally (pp.153-154). The basic premise of this emotional contagion hypothesized process then is that people tend to mimic others physical movements, and the newly assumed gestures, postures, facial expressions, or their combination, become feedback that then influences internal emotions. Hatfield et al. (1994) describe how quickly the sequence of events can occur in the emotional contagion process. Reaction to emotions, measured in milliseconds, have been shown to be quicker than the time it takes to cognitively process the information and make a conscious decision to mimic the behavior (Davis, 1985). The important thing to stress here is the speed with which the emotional contagion process occurs can rule out the possibility of conscious cognitive mechanisms. Emotions internal to the sender are exposed through various physical outlets, such as facial expression, body posture, hand gestures, and vocal style (Hatfield et al., 1994). Such stimuli are accessible to conversant awareness, while other stimuli, such as facial muscle movements, may be so subtle that only electronic devices can detect them (Cacioppo & Petty, 1983). Once the senders emotions generate an outwardly observable physical behavior, the observer witnesses the behavior, and immediately and unconsciously mimics the behavior. Indeed, Hatfield et al. (1994) suggest that individuals


39 may constantly scan the face during a conversation to gather additional emotional information. Once the observer has mimicked the facial expression, a similar emotion is generated internal to the observer. Although the feedback process from the mimicked facial expression to the resulting emotions remains unclear, researchers agree that muscular movements and emotional states are reciprocally linked (Adelmann & Zajonc, 1989). Subjects asked to frown are angrier than those asked to smile, and in addition, subjects have difficulty feeling emotions that conflict with their facial expression (Laird, 1984). Provine demonstrated that both yawning (1986) and laughing (1992) can stimulate mimicry. Regardless of our intentions, when we see someone yawning, not only might we yawn, but also become tired. Likewise, if we observe someone laughing, we might smile or laugh, and become happy simply because we witnessed and subsequently mimicked another who has initiated the behavior. It is difficult to consciously adopt an emotion that is incompatible with the expression being imitated (Hatfield et al., 1994). This suggests that as much as we might try to avoid catching someone elses emotion, it is difficult for us to feel an emotion that is inconsistent with our facial expression or posture. This non-voluntary and subconscious aspect of emotional contagion is intriguing in that every one is susceptible to it to some degree. Thus, empirical data have generally supported the existence of emotional contagion based on psychological and physiological mechanisms that are often unobservable (Hatfield et al., 1994). In sum, the key component to the emotional contagion mechanism which allows a transfer between two individuals is mimicry. Individuals that feel certain emotions express these emotions in their faces. Other individuals mimic their facial expressions


40 and than through the physiological feedback enacted by these facial expressions they start feeling the same corresponding emotions. Thus, if you feel happy you express it in your face (i.e., smile) other people mimic this facial expression (they also smile) and as a result they start feeling happy themselves. Applying this reasoning to the leader-follower dyads, it becomes evident that leaders can literally be contagious carriers of emotion who expose followers to this emotion. There is some evidence to suggest that there is an interaction between mimicry and the importance of the agent mimicked. In other words, mimicry may be stronger when the individual being observed is considered by the observer to be in a position of power (Hatfield et al., 1994). Consequently, because of their position power and authority, leaders are likely to be considered more important, resulting in greater levels of mimicry among observant followers. One study particularly generated support for this emotional transfer hypothesis of happiness between leader and followers. McHugo, Lanzetta, Sullivan, Masters, and Englis (1985) found that observers of a President Reagan speech shared his happiness or anger, regardless of whether they were supporters or opponents of the president. A particularly interesting aspect of this study is that the opponents of Reagan claimed to have had an overall negative reaction to Reagans emotional displays, but their recorded facial expressions and skin resistance levels indicated happiness during positive displays of emotion by him. This result suggests that emotional contagion may occur between leader and follower even if the leader is not particularly liked by the follower, attesting to the strength of this sub-conscious process. In sum, due to their inherently prominent positions in organizations, leaders are naturally more conspicuous than followers. For this reason, followers most likely pay


41 more attention to their leader behaviors and gestures. Followers observe the leaders emotional expressions, subconsciously mimic these expressions, and ultimately feel the same emotions as the leaders. Consequently, leaders who are satisfied will be happier (Judge & Hulin, 1993; Judge & Locke, 1993) and should express emotions congruent with their happiness, while followers should observe and mimic the leaders happy expressions, resulting in increased happiness for followers (Hatfield et al., 1994). Finally, according to the recursive relationship suggested by Judge and Locke (1993), induced happiness in followers should increase follower satisfaction as well. However, leaders are not always easy to observe, especially in distal leader-follower dyads. In this case, when followers cannot directly observe leaders, the mimicry cannot take place. Although studies show that mimicry can take place between distal charismatic leaders and followers (McHugo et al., 1985), the lack of contact with distal leaders cannot explain long term sustained contagion. In contrast, proximal leaders in direct contact with followers are more easily observed by followers, thus increasing the likelihood of mimicry. Followers who work in close proximity or in close interaction with leaders are exposed to leaders emotion expressions throughout the day, and should catch the leaders emotions. For example, a leader who walks through the office with a smile will tend to generate more smiles among followers, while a leader who walks around with a frown or scowl, will generate similar negative expressions in the followers. More specifically, charismatic leaders who tend to display positive emotion and energy (Conger & Kanungo, 1987), will tend to spread this optimism among the followers. Social learning theory (Bandura, 1973) suggests that many emotional responses are a direct result of social conditioning. Considering that followers who are close to the


42 charismatic leader can observe the follower frequently throughout the day, it is likely that these followers will have sustained exposure to the leaders emotions, resulting in sustained attitudes that are influence by associated emotions. Social Information Processing In addition to the previously discussed job satisfaction models (Locke, 1976; Hackman & Oldman, 1976), Salancik and Pfeffer (1978) proposed another mechanism for explaining the development of employee attitudes and behaviors. As opposed to the former satisfaction models which rely on individual personal characteristics to define needs and values, the social information process suggests that an individuals needs and values are not defined independently by the individual, but are also influenced by coworkers as well as organizational climate (Salancik & Pfeffer, 1978). Social information theory suggests that the social environment provides cues that individuals use to construct and interpret events and personal needs, and also emphasizes the relative salience of information (Salancik & Pfeffer, 1978). In other words, employees attitudes and interpretation of events are influenced by coworker statements reflecting the coworkers own attitudes and interpretation of events. For instance, a coworkers comment suggesting that he needs a raise can influence other employees as they evaluate their own monetary needs, possibly resulting in a discrepancy between needs and wants via a social comparison process (Festinger, 1954). Likewise, a coworkers comment about the ineptitude of a newly hired supervisor can influence other employees impression of the new supervisor. While it is difficult to distinguish the effects of an employees personal characteristics on individual attitudes from the effects of social information on individual attitudes (Salancik & Pfeffer, 1978), research indicates that group affiliations (Herman & Hulin, 1972) and organizational characteristics (OReilly &


43 Roberts, 1975) can have a greater influence on attitudes than individuals. These results support the social information theory and suggest that individual attitudes such as satisfaction are indeed influenced by other individuals in the organization. Social information processes not only include information obtained by coworkers, but also from leaders, since both are also inherently members of the social environment within an organization. Leaders, by the nature of their organizational positions, are required to communicate with their followers to provide task definition, task prioritizations, performance feedback, etc. Not only do leaders provide the bare facts necessary for followers to perform their jobs (hopefully), but leaders are also likely to provide solicited and unsolicited comments containing a wide range of content, from helpful supplemental job-related information to extraneous quips, all of which may contain bits and pieces of the leaders values, preferences, attitudes, expectations, or perceptions that followers piece together as they construct their environment (Weick, 1977). However, considering that leaders possess authority and position power over followers, followers are likely to not only pay more attention to information originating from the leader (Hatfield et al., 1994), but may also assign more weight to the information. Leaders in close proximity to followers will provide more frequent information about behavior and attitudes, and over time, the continuous displays among followers can lead to general alignment of follower attitudes (Salancik & Pfeffer, 1978). Thus, the specific relationship between leader satisfaction and follower satisfaction can also be explained by the social information processing theory. As discussed previously, individuals satisfied with their jobs tend to be happier (Judge & Locke, 1993), and as such are likely to openly display behaviors and emotions consistent with their


44 internal happiness (Hatfield et al., 1994). Furthermore, individuals who are happy may tend to verbally communicate to other individuals such things as the cause of their happiness or the intensity of their happiness. The content of the information communicated can have an influence on other individuals. For example, a satisfied and therefore happy employee arrives at work and publicly pronounces over the top of the cubical walls, This is gonna be a great day! Other employees who are within earshot hear this proclamation and reevaluate their outlook on the day by decreasing their expected value and/or increasing the value provided by the job, thus diminishing the discrepancy between the two and increasing overall satisfaction. Furthermore, if a leader were to pronounce the same exact statement over the cubical walls, it is likely to have a greater effect on the followers attitudes simply because of the position power the leader inherently possesses. If a leader is satisfied, it is likely that followers will adapt their own values and attitudes to align with the leaders attitudes concerning satisfaction. Using this line of reasoning, the social information process may be a possible mechanism for the transfer of job satisfaction from the leader to the follower. Thus, both the emotional contagion hypothesis and the social information hypothesis suggest that H3: Leader satisfaction will be positively related to follower satisfaction. Summary of Research Proposal and Mediation Hypotheses As previously suggested, the assumption of the proximal leader-follower dyad is not necessarily valid in large organizations where top-level leaders are far removed from their followers, and as such, other charisma transfer mechanisms must be proposed to explain the effects of top-level leader charisma on bottom-level follower satisfaction. One of the propositions presented in this dissertation is that intermediate-level leaders act as conduits to facilitate and mediate the charismatic influence of top-level leaders as it


45 cascades down through the organization. The previous discussion provided theoretical support for the links on which these two multilevel paths rely. First, the argument was made that leader charisma can influence follower satisfaction by utilizing previously established job satisfaction models such as the value-percept model and the job characteristics model. Next, the argument was made that leader charisma can also influence follower charisma through affective processes, role modeling, identification and halo effects due to categorization. Finally, the argument was made that leader satisfaction can influence follower satisfaction through emotional contagion and social information processes. The establishment of these three relationships is critical in the development of the two multilevel models. The multilevel models propose a series of two links across three levels: between the level-3 (i.e., top-level) leader and level-2 (i.e., intermediate-level) leader, and then between the level-2 leader and the level-1 (bottom-level) follower. Since two intermediate-level leader variables were measured, two possible multilevel paths can be tested (see Figure 2-1). The first involves leader charisma at level-3, leader satisfaction at level-2, and follower satisfaction at level-1. This path proposes that top-level leader charisma influences intermediate-level leader satisfaction, and that this intermediate-level leader satisfaction in turn influences bottom-level follower satisfaction. As indicated in Figure 1, this multilevel model requires that the intermediate-level leaders satisfaction mediates the influence of the top-level leaders charisma on bottom-level followers satisfaction. As previously mentioned, each of the two individual links in this multilevel model can be sufficiently explained using existing theory.


46 The second multilevel path involves leader charisma at level-3, leader charisma at level-2 charisma, and follower satisfaction at level-1. The only difference in this multilevel path is the mediating variable at level 2. In this case, instead of using intermediate-level leader satisfaction, this path relies on intermediate-level leader charisma to mediate the influence of the top-level leaders charisma on bottom-level followers satisfaction. H4a: Intermediate-level leader satisfaction mediates the relationship between top-level leader charisma and bottom-level follower satisfaction H4b: Intermediate-level leader charisma mediates the relationship between top-level leader charisma and bottom-level follower satisfaction


CHAPTER 3 STUDY 1 Method Participants The population from which the sample was drawn was a major southeastern U.S. metropolitan fire department. The total sample size of the fire stations was 273 with 216 bottom-level fire-fighters, 48 intermediate-level officers, and 9 district chiefs. The average age within the sample was 34.4 years, with only 5% of the sample being female. Ethnically, the sample was diversified as such: 76% white, 13% Hispanic, 10% Black, and 1% other. The entire city fire department consists of four districts, but only three of them were sampled. One district had been recently established as a result of an internal reorganization and was not sampled due to its unique situation. Each district consists of at least three fire stations dispersed within relatively close physical proximity of each other and adequately sized to meet the fire protection requirements in each stations zone of responsibility. In those districts with more than three stations, a convenience sample was taken based on location. A district chief is in charge of all three shifts at each fire station in the district. A fire station contains from one to five fire engines or other similar fire protection vehicles--the mix of vehicles depends on the types of structures existing in the area of responsibility. Each fire fighting vehicle is led by an officer, who is responsible for supervising from three to five fire fighters or paramedics assigned to that vehicle. 47


48 Although the hierarchical organization of the physical assets (i.e., district, station, and engine) is in the typical pyramid format, the hierarchical organization of the personnel (district chief, officer, and firefighter) is much flatter since there no overall station leader. Instead, each officer within a district reports directly to his respective district chief. As a result, a district chief can supervise up to eight officers in three different stations (see Figure 3). Furthermore, since each station has three shifts, the district chief supervises three identical organizations. Because of this unique organizational structure, an assumption was made about the leaders which allowed an increased sample size of level-3 leaders. Since the three shifts at any one station work independently of each other, and have relatively little interaction, each shift was considered a separate organization and was assumed to evaluate the district chief independently of the other shifts. The district chief was rated by each of the members in each of the three shifts, resulting in three different leadership scores. As such, instead of only having nine top-level leaders, the assumption resulted in a sample size of 27 top-level leaders. Figure 3-1. Typical fire department organizational structure


49 The fire fighter duty schedule is based on 24-hour shifts, which rotates among three teams labeled A-shift, B-shift, and C-shift. Each team resides at their station for the duration of their 24 hour shift. This schedule allows each team 48 hours off-duty between shifts. The team essentially lives together in relatively close quarters where they eat family-style meals and spend free time in recreation or exercise rooms. The typical day can involve dozens of calls such as personal injuries, vehicular accidents, fire alarm activations, and fires. Inherent in each of these daily events is some degree of personal risk for the fire fighters, from minor injury to possible death. Consequently, fire fighters rely heavily on teamwork among their fellow fire fighters to reduce or manage this risk. Procedure The fire fighters were visited personally by the researcher in their stations for the administration of a paper survey. The fire fighters were given a short explanation of the purpose of the survey; they were told it was a survey asking about leadership and personality differences. Prior to starting the survey, the researcher informed the subjects that in the event of a fire alarm to which the fire fighters must respond, the researcher would collect all surveys and allow the subjects to complete the survey upon their return. The researcher returned to each station as many times as necessary to administer the survey to each fire fighter. Only one subject refused to participate. Measures Leader charisma. Leader charisma was measured by the charisma sub-scale of the Multi-Factor Leadership Questionnaire (MLQ; Bass & Avolio, 1990). Thus, of the 45 items of the MLQ, twelve items were used representing three dimensions of charisma: idealized influence (attribute), idealized influence (behavior), and inspirational motivation. The members of each fire engine (excluding the leader) were asked to rate


50 the extent to which such items as "I have complete confidence in him/her" and "Is a model for me to follow" characterized the designated leader of their fire engine. Similarly, the members of each district were asked to rate these items to the extent that they represented the district chief. The coefficient alpha reliability of the scale in this study was = .96 for the fire engine officer and .96 for the district chief. A statistically significant one-way random effects ICC (.17) indicated support for aggregation across raters at the fire engine level (see Bliese, 2000 for review). The ICC(.16) for the district level also indicated support for aggregation across followers. Work group satisfaction. The dependant variable in this study was fire fighters satisfaction with their work group. The fire fighter satisfaction measure was based on LePine and Van Dynes (1998) five-item, seven-point faces scale. LePine and Van Dynes scale used the graphics from Smith, Kendall, and Hulin (1969) measure. Because the graphic faces were outdated and based on the male gender only, a modified version was used, similar to the gender-neutral faces used by Kunin (1955). The items asked participants to select the face that best expressed how they felt about their work team, the members of the team, the quality of interaction among team members, the information they got from team members, and the influence they had with their team. Coefficient alpha reliability estimate was .79 for bottom level followers, and .85 for officers. Analysis The data of this study were hierarchical in structure with three levels of analyses such that the dependent variables were members satisfaction with their work groups at the followers-level with fire fighters grouped into fire engines and fire engines grouped into districts. In order to analyze these data, I drew on recent advances in multilevel


51 statistical theory by using the multilevel regression techniques of hierarchical linear modeling (HLM, Byrk and Raudenbush, 1992). Hierarchical linear modeling involves simultaneously estimating several regression equations for the dependent measure. At the first level of analysis (i.e., the fire engine-level model), I specified for the individual fire fighters in the data set the following model: Level 1: ijk = 0jk + eijk where ij was the satisfaction of a firer fighter i with his work group, at fire engine j in district k; 0jk (the intercept) represented the mean satisfaction of fire engine j members in district k; and eijk represented the individual error term, that is the deviation of member ijk from his or her fire engine score. In the second-level model (i.e., fire engine-level model), I tested two different models. While the dependent variable was the intercept from Level 1 model, the independent variable (X0jk) was either the intermediate-level leader (e.g., officer) level of charisma or level of satisfaction. Thus, Level 2: 0jk = 00k + 01kX0jk + r0jk where 00k (i.e., level-2 intercept) represented the mean satisfaction of officer k, and 01k was the slope of the regression line representing the strength and direction of association between the level-2 predictor variable, X0jk (officer charisma or satisfaction), and 0jk; r0jk represented the level-2 random error that captured the deviation of 0jk (mean fire engine j satisfaction) from the district mean level of satisfaction. At the third level model (i.e., district level model), I tested the influence of Level-3 leaders charisma (i.e., district chief) on the intercept (00k) of Level-2. Thus,


52 Level 3: 00k = 000 + 001W01j + u00k 01k = 010 + u01k where 000 and 010 were intercepts for the third-level models (district level) for 00k and 01k, respectively; W01j represented the level-3 predictor variable, which was district chief charisma. This level-3 variable predicted the level-2 intercept; u00k and u01k were the level-3 random effects that represent the deviations of actual 00k and 01k, respectively, from their predicted values. All HLM coefficients were uncentered. In HLM all levels are tested simultaneously and therefore the combined model that was tested was: ijk= 000 + 001W01j + 010X0jk + u00k + u01k+ r0jk+ eijk HLM coefficients were standardized by multiplying by the standard deviation of each predictor, respectively, and dividing by the standard deviation of the outcome variable, which converts the coefficients to standard deviation units (Hox, 2002). Such a procedure removes the effects of instrument scaling and allows one to observe the strength of the relationship relative to the variance of the measure. Results The main hypothesis of this study was that top-level leader charisma cascades through the leadership ranks to influence bottom level followers satisfaction. In other words, I hypothesized that top-level leaders charisma would affect followers satisfaction, and that these relationships will be mediated by intermediate-level leader variables. In order to test this hypothesis I developed five HLM models. The description of these models and the results of the HLM analyses testing these models are presented in


53 this section. Table 1 gives a tabular summary of the multilevel relationships sequentially tested. Table 3-1. Hierarchical linear model summary Sequence of Models Hierarchical Level 1 2a 2b 3a 3b 3 Charisma Charisma Charisma Charisma Charisma 2 Satisfaction Charisma Satisfaction Charisma 1 Satisfaction Satisfaction Satisfaction Model 1 The first step in a mediation analysis (see Baron & Kenny, 1986) is to determine the existence of a relationship between the main predictor, district chief (i.e., level-3 leader) charisma, and the dependent variable, fire fighter (i.e., level-1 follower) satisfaction, across the multiple hierarchical levels. Accordingly, in Model 1 I tested a three-level hierarchical linear model where I analyzed the effects of top-level leader charisma (X0jk) on the work group satisfaction of bottom level (i.e., level-1) followers. Even though a level-2 variable was not needed in this model, it was still necessary to use a three-level model with an empty level-2 to distinguish the variance attributed to each level. The HLM results for Model 1 are presented in Table 3-2 and show that district chief (i.e., level 3 leader) charisma positively and significantly (=.23, p < .05) influenced follower satisfaction among fire fighters (i.e., level 1 followers), thus supporting the first hypothesis.


54 Once the relationship between top-leve l leader charisma and bottom-level satisfaction was established, I tested how district chief ch arisma cascaded downward to influence fire fighters satisfac tion. In other words, I tested the possibility that top-level leader charisma effects are mediated through intermediate-lev el leader (officer) variables at level 2. Thus, I examined possible paths fr om level 3 to level 1 via the two suggested mediating variables (charisma and satisfaction) at level 2. Table 3-2. Model 1: three-level model of th e influence of level-3 leader charisma on level-1 follower satisfaction, Study 1. Fire Fighter Satisfaction Raw Standardized Intercept 00k 4.10 ** (.99) --Leader (District Chief) Charisma 01k .45* (.24) .23 Note : ** p < .01, p < .05. Raw refers to raw coefficients. Standard errors are in parentheses. Models 2a and 2b Models 2a and 2b investigated the sec ond step of the mediation by testing the effects of level-3 charisma on the two level-2 variables, intermediate-level leader (officer) satisfaction and intermediate-level leader charisma, respectively. Table 3-3 presents the results of these two models and sh ows that district chief (i.e., level 3 leader) leader charisma did not significantly influence ( =-.04, ns ) officer (i.e., level 2 leader) satisfaction. Thus, H1 was not supported here with regard to the charisma-satisfaction link between upper-level leader s and intermediate-level leaders. As Model 2b shows,


55 district chiefs charisma did significantly influence officers charisma (=.45, p < .05), supporting H2. Table 3-3. Two-level models of the influence of level-3 leader charisma on level-2 intermediate leader satisfaction (model 2a) and charisma (model 2b), Study 1. Note: **p < .01, *p < .05. Raw refers to raw coefficients. Standard errors are in parentheses. Model 2a Officer Satisfaction Model 2b Officer Charisma Raw Standardized Raw Standardized Intercept 00k 6.04** (1.39) --1.60** (.92) --Level-3 Leader (District Chief) Charisma 01k -.04 (.35) --.61* (.22) .45 Models 3a and 3b Models 3a and 3b tested the third step in the mediation analysis by investigating the effects of the two intermediate-level predictors on bottom level follower satisfaction, controlling for upper-level predictor. Thus, models 3a and 3b examine the effects of officer (i.e., level 2) satisfaction and charisma, respectively, on bottom-fire fighter (i.e., level 1 follower) satisfaction, controlling for upper-level (district chief) charisma. The results of these models are presented in Table 3-4. Model 3a shows that officer satisfaction was not significantly related to fire fighter satisfaction (B=.09, ns ). Thus, the third hypothesis was not supported here. Because both links from upper-level leader charisma to intermediate-level leader satisfaction and from intermediate-level leader


56 satisfaction to follower satisfaction were not significant, the mediation process via level 2 satisfaction was not supported. Model 3b showed that officer charisma significantly (=.13, p < .05) influenced fire fighter satisfaction, once again supporting the first hypothesis. Because both paths from upper-level leader charisma to intermediate-level leader charisma (Model 2b) and from intermediate-level leader charisma to follower satisfaction (Model 3b) were significant, mediation is possible. Indeed, as Table 4 shows, when intermediate-level charisma was added to the model, the relationship between upper-level leader charisma and follower satisfaction that originally was significant (B=.45, p<.05) dropped to an insignificant level (B=.29), indicating full mediation. Table 3-4. Three-level models of the influence of level-2 intermediate leader satisfaction (model 3a) and charisma (model 3b) on level-1 follower satisfaction, Study 1. Model 3b Satisfaction Mediating Model 3b Charisma Mediating Raw Standardized Raw Standardized Intercept 000 4.05** (.84) --4.00** (.83) --Level-3 (District Chief) Leader Charisma 001 .33* (.17) .33 .29 (.17) --Level-2 (Officer) Leader Charisma 010 .09 (.08) --.19* (.11) .13 Note: **p < .01, *p < .05. Raw refers to raw coefficients. Standard errors are in parentheses. One possible explanation for the non-significant path from upper-level leader charisma to follower satisfaction through intermediate-level leader satisfaction was small sample size of leaders. There were 27 leaders at the top level and only 48 leaders at the


57 intermediately-level which is a ratio of less than two on average. That may have made the results of the HLM analysis unstable. However, this problem was mitigated in Study 2, where the number of participants, especially the number of leaders, was significantly larger.


CHAPTER 4 STUDY 2 Method Participants The population sampled in this study was the cadet student body at the United States Air Force Academy (USAFA), located in Colorado Springs, Colorado. The sample size included 974 cadets, ranging in age from 17 to 25 with an average age of 19.8; 77.5% were males, and 88.2% were white. More specifically, at each of the three hierarchical levels from top to bottom the average ages were 21.9 years (n=25), 21.5 years (n=64), and 19.6 (n=855). USAFA is the primary officer commissioning source whose mission is to recruit and develop future Air Force leaders. The four year Bachelor of Science curriculum is supplemented with weekly seminars and training on leadership. Consequently, some refer to USAFA as a leadership laboratory. The student body at USAFA is known as the cadet wing and consists of approximately 4000 cadets from all 50 states, the territories and several foreign countries. One additional unique characteristic of this population is the fact that cadets are initially assigned to squadrons based on a wide range of demographic variables and athletic and academic aptitudes in an effort to evenly distribute and control for such variables across squadrons. The uniformly structured organizational units as well as the administrative efforts to prevent unequal distribution of any one variable makes the cadet population idea for such a study. 58


59 The cadet wing is subdivided into four groups (i.e., group 1 to group 4), and each group is further subdivided into nine squadrons, resulting in a total of 36 squadrons in the entire cadet wing. The typical squadron size is approximately 110 cadets. Cadets at all levels in the squadron organization are delegated specific authorities and responsibilities through formal chains of command. In general, the cadets are in charge of running the daily squadron operations and all required duties. However, each cadet squadron is supervised by an Air Officer Commander (AOC) who is an active duty, commissioned officer with command authority over the cadets who provides instruction, and serves as a role model as the cadets experience firsthand the processes of leadership and organization to accomplish the mission. Although cadet squadrons are essentially self-governing, each must possess an identical organizational structure consisting of four distinct hierarchical levels, which is ideal as a sample for multilevel research. Each cadet squadron is subdivided into three main flights (i.e., A, B, C), each being led by a flight commander. (A fourth, smaller administrative flight also exists, consisting of cadets in staff positions within the squadron. Since it is different in structure and membership, it was not considered in the study.) The top-level leader in this study was the flight commander. While it would have been ideal from a theoretical standpoint to focus on squadron-level leaders, the limited number of squadron commanders combined with the expected response rate would have resulted in a very small sample size. Consequently, this study focused on leadership from the flight commander level and below. The element is the smallest organizational unit in the cadet squadron, and each flight is divided into three elements (1, 2, 3), each led by an element leader. The


60 approximate number of cadets in an element is 12, resulting in a typical flight size of about 36 cadets. The element leader was the intermediate-level leader for the purpose of this study, and all other cadets within each element were the bottom-level followers. Figure 4-1 shows the organizational chart for a typical cadet flight. The total numbers of flight commanders, element leaders, and followers in the entire cadet wing are 108, 324, and approximately 3600, respectively. Although this entire cadet wing population is relatively large, the salient sample size in this study is the number of leader responses at the flight commander level. Since flight leader charisma was measured by aggregated follower ratings, the available data points for the HLM analysis at the flight level (i.e., level-3) was 108, since enough followers rated each flight commander, regardless of whether or not the specific flight commander responded to the survey. However, the same luxury was not available at the element leader level (i.e., level-2), since the other level-2 variable, satisfaction, was a self-report measure. Therefore, the number of level-2 data points for the HLM analysis was 64, which was the actual number of level-2 respondents. These 64 level-2 leaders resided in only 54 of the 108 possible flights, resulting in a level-3 sample size of 54 for the HLM analysis. FlightCommander Element 1Leader 10-12 Cadets Element 3Leader 10-12 Cadets Element 2Leader 10-12 Cadets Top-level Leader (i.e., Level 3) Intermediate-level Leaders (Level 2) Bottom-level Followers (Level 1) Figure 4-1. Typical cadet flight structure


61 Procedure The survey was administered electronically via an Internet website. The entire cadet wing was first invited to participate via an email from a senior officer from the Commandant of Cadets office. The email contained a brief overview of the study and a link to the Internet website survey. The survey began with questions that filtered out subjects who did not belong to the target organizational unit. Since the target population in this study was only leaders and followers within cadet flights, responses from cadets above or outside of the flight unit would not be useful. These respondents were directed to a Thank You page. Once the respondents indicated that they were members of the target population, they were prompted with information about the study and use of data, all of which were required by USAFA prior to obtaining the respondents informed consent. The previously described measures were administered to the respondents, with each web page containing questions for that specific measure. After the measures were administered, the respondents were asked standard demographic questions. The number of cadets that elected to participate in the survey was 974 which represented a response rate of 27.22%. Measures Leader charisma. Leader charisma was measured by the same charisma sub-scale of the MLQ (Bass & Avolio, 2000) used in Study 1. The coefficient alpha reliability of the scale in this study was = .98 for the element leaders and .97 for the flight leaders. A statistically significant one-way random effects ICC (.12) indicated support for aggregation across raters at the elements level (see Bliese, 2000 for review). The ICC (.10) for the flights level also indicated support for aggregation across followers at the flight level.


62 Work group satisfaction. Satisfaction with work group was measured in this study by the same scale used in study 1. Coefficient alpha reliability estimate was .83 for bottom level cadets, and .82 for element leaders. Results Model 1 The analysis used in study 2 replicated that of study 1 with the same HLM models. Model 1 tested the effects of level-3 leader charisma on level-1 follower satisfaction. The HLM results for Model 1 shown in Table 4-1 indicate that flight leader charisma at level 3 (i.e., group level) positively and significantly influences follower (cadet) satisfaction at level 1 (=.27, p < .01). Similar to Study 1, this result supports the first hypothesis indicating that distant leaders influence bottom level followers. Thus, next I tested the mediation process. Table 4-1. Model 1: three level model of the influence of level-3 leader charisma on level-1 follower satisfaction, Study2. Cadets Satisfaction Raw Standardized Intercept 00k 2.79** (.25) --Level-3 Flight Leader Charisma 01k .58** (.09) .27 Note: **p < .01, *p < .05. Raw refers to raw coefficients. Standard errors are in parentheses.


63 Models 2a and 2b As shown in Table 4-2, similar to Study 1, Model 2b, level-3 leader (flight) charisma had a positive and statistically significant effect on level-2 element leader charisma (=.16, p < .01), supporting the second hypothesis. In contrast to the Study 1 findings, in this study the relationship with level-2 leader satisfaction was statistically significant (=.26, p < .05) suggesting that upper-level leaders did in fact affect intermediate-level leader satisfaction. Thus, H1 was supported also with regard to intermediate-level leader satisfaction. Table 4-2. Two-level models of the influence of level-3 leader charisma on level-2 intermediate leader satisfaction (model 2a) and charisma (model 2b), Study 2. Note: **p < .01, *p < .05. Raw refers to raw coefficients. Standard errors are in parentheses. Cadet Satisfaction (Model 2a) Cadet Charisma (Model 2b) Raw Standardized Raw Standardized Intercept 00k 3.78** (.69) --2.36** (.28) --Level-3 Flight Leader Charisma 01k .48* (.23) .26 .28** (.10) .16 Models 3a and 3b Here again, as in Study 1, Model 3a showed that the relationship between intermediate-level leader satisfaction and follower satisfaction was not significant, controlling for upper-level leaders charisma. Thus, H3 was also disconfirmed in this


64 study. Because the path from intermediate-level leader charisma to follower satisfaction was not significant, it is not likely that mediation has occurred here. Thus, H4 has also been disconfirmed in this study. The results of model 3b are represented in Table 4-3, indicating that level-2 leader (element leader) charisma had a positive and statistically significant relationship with level-1 follower (cadet) satisfaction (=.40, p < .01), controlling for upper-level leader charisma, again supporting H1. Here again, the significance of the two paths from upper-level leader charisma to intermediate-level leader charisma and from intermediate-level leader charisma to follower satisfaction indicate a possible mediation. Table 4-3. Three-level models of the influence of level-2 intermediate leader satisfaction (model 3a) and charisma (model 3b) on level-1 follower satisfaction, Study 2. Model 3a Satisfaction Mediating Model 3b Charisma Mediating Raw Standardized Raw Standardized Intercept 000 2.30** (.52) --1.36* (.56) --Level-3 Flight Leader charisma 001 .65** (.17) .58** (.14) Level-2 Element Leader satisfaction 010 .05 (.08) --.43** (.12) .40 Note: **p < .01, *p < .05. Raw refers to raw coefficients. Standard errors are in parentheses. In contrast to Study 1, here the coefficient from upper-level charisma to follower satisfaction did not drop to an insignificant level when intermediate-level leader charisma was added to the model. This indicates a partial mediation effect.


CHAPTER 5 DISCUSSION Summary of Results While the extant literature indicates that leader charisma can influence follower satisfaction, the process linking these constructs remains in doubt. Furthermore, most existing charismatic leadership theories rely on direct dyadic relationships between leaders and followers, an assumption that becomes increasingly problematic as organization size increases. Considering that such a large majority of charismatic leadership literature relies on the dyadic-direct structure to explain the transmission of charisma, charismatic leadership research seems to be greatly limited in scope. This dissertation instead, proposes alternative mediating processes to explain cascading leader charisma through an organization. The hypotheses introduced in this dissertation suggest that leader charisma influences both intermediate-level leader charisma and satisfaction, which in turn influence follower satisfaction. The main mediation models used in this dissertation (see Figure 2-1) were developed to explain the path by which leader charisma travels down the organization hierarchy to reach followers in the absence of direct and proximal leader-follower dyads. Although testing these mediation models was the ultimate objective, a series of tests of smaller models was required to verify the existence of relationships among the proposed components on which the main mediation models rely. In study 1, the results from this series of five HLM models indicated that in two of the models (1 & 3b) leader charisma can influence follower satisfaction. Upper-level leader charisma was also related to 65


66 intermediate-level leader charisma (Model 2b). However, the upper-level leader charisma was not related to intermediate-level leader satisfaction (Model 2a) nor was intermediate-level level satisfaction related to follower satisfaction (Model 3a). I found mixed results in the mediation model from upper-level leader charisma to follower satisfaction through intermediate-level leader charisma. The path through intermediate-level charisma seemed to be significant, while the other path through satisfaction did not appear to indicate significant results. Study 2, which consisted of a larger sample size and significantly more leaders at the intermediate level than Study 1, produced more supporting evidence to the some of the hypotheses of this study. In this study four of the five models produced significant results. Especially important was the mediation models 3b that showed that cadets at the lower level were significantly influenced by flight level leaders charisma and this influence could be partially explained by the charisma of element leaders (i.e., intermediate-level leaders). In other words, the more charismatic was the flight leader, the more charismatic were the element leaders, and this explained the satisfaction of the cadets. The robust results from study 2 then, which utilized a larger sample size, are very encouraging in the effort to identify mediating processes for leader charisma. A New Approach to Charismatic Leadership The ideas presented in this dissertation introduce a new approach to the charismatic leadership process. While extant charismatic theories are overly constrained by the direct dyad assumption, the results of this dissertation suggest that alternative hypotheses relying on mediation processes may more accurately model the charismatic leadership process. The amount of direct contact that leaders have with followers is indirectly related to the size of the organization. In large organizations, it is impossible for leaders


67 to make regular contact, or even any contact at all, with every member in the organization. This might suggest that all leaders are handicapped by the size of their organizations, and that they can only expect to influence a small minority of the people directly related to them in their pursuit of organizational goals. Regardless of the intensity of any single leaders charisma, in large organizations it is unrealistic to expect any lasting impact on followers by means of charismatic messages delivered on a sporadic basis. How then does a leaders valuable charismatic influence reach all the members of an organization to ignite the fire as researchers have suggested (Klein & House, 1995)? Researchers and layman alike intuitively suspect that leader charisma trickles down through an organization, but without a theory to explain the process, any such discussion is simply hearsay. Followers require more frequent reinforcement in order to sustain the long term energy, enthusiasm, and motivation that leaders so desperately desire. Research shows that individuals attitudes and moods are quite stable, meaning that although employees attitudes and moods can be temporarily suspended by specific events or manipulations, individuals eventually return to their normal state (Staw et al., 1986). This tendency can even be shown with political views that are usually quite strongly held by people. For example, one study tested the effects of a four year education at a politically liberal institution on students who were politically conservative upon enrollment; the results showed that the conservative students adopted liberal attitudes during the four-year period, but returned to their conservative values shortly after leaving (Necomb, 1943). If the habituation process can be observed in individuals even after a four year hiatus from


68 their normal attitudes, it is very likely that the same process can be observed when the manipulation period is drastically reduced to much shorter time spans. Applying this habituation theory to charismatic leadership means that while infrequent interjections of charisma by a distant leader may temporarily spike the attitudes or moods of followers, the followers quickly return to their normal state. This further exacerbates the problem associated with the direct dyad assumption. If followers need consistent reinforcement to sustain attitudes, in large organizations they are most likely not getting it through direct dyadic contact with the charismatic top-level leader. Since most followers in any large organization have infrequent direct contact with the charismatic top-level leader, the few encounters followers experience with the charismatic leader most likely induces an infinitesimally short manipulation period. Such limited direct contact can only hope to affect short term attitudes versus the long term ones. Nonetheless, leader charisma still seems to provide an observable lasting influence on followers at all levels. Other processes, such as mediation of charisma by intermediate-level leaders, must exist than to explain the influence of top-level leader charisma. The results of this research indicate that the influence of leader charisma is not necessarily constrained by proximal and direct dyadic relationships or by the size of the organization. Instead, this research suggests that a leaders charisma can cascade down through an organization to bottom-level followers via intermediate-level leaders. Consequently, charisma acts partially through others, thus revealing the importance of intermediaries along the chain of command. The results presented in this dissertation indicate that the relationship between top-level leader charisma and bottom-level follower


69 is partially mediated by intermediate-level leaders who act as charisma conduits. The models used to test and support this proposition show that leader charisma can influence both charisma and satisfaction in direct subordinates, which in turn, influences satisfaction in followers. Thus, intermediaries are an important part of the charisma transfer process. Some researchers have suggested that charisma is a socially constructed phenomenon that depends more on social information than on the actual charisma of the leader (Hogg, 2001). This line of research also recognizes the flaws associated with the direct dyad assumption and as a result suggests instead that followers use implicit leadership theories to categorize leaders versus cognitively analyzing leader ability (Lord et al., 1984). One criticism of this theory is that it seems to discount charismatic behavior displayed by the leader, and instead focuses on the information obtained in the social environment regardless of whether it is accurate or not. In other words, this theory suggests that often followers appraisals of their leaders charisma can be completely off the mark, thus emphasizing the diminution of the leaders true charisma. This perception is very likely wrong given the accumulating evidence in favor of charismatic theories. That is not to say that the social construction theory is not without merit, however. Considering the potentially important role played by intermediate-level leaders in the transfer of charisma, we must also recognize that these intermediaries are likely to provide key information on which followers socially construct their impression of the leaders. Practical Applications As we digest the realization that intermediaries play a pivotal role in the charisma process, it introduces two practical questions deserving of discussion. The first and most


70 obvious question is what happens when the intermediaries are bad conduits of charisma? Do such intermediaries create dead ends for a leaders charismatic influence, resulting in pockets within the organization that are deprived of upper-level leadership influence? Since this research indicates that intermediate-level leaders can transmit top-level leader charisma, it is important that the intermediate-level leaders are cognizant of their responsibility to act as charismatic conduits in conveying their leaders vision, goals, etc., down to the followers. On the other hand, when top-level leaders recognize a lack of ability of their intermediate-level leaders to transmit charisma, it may require the top-level leaders to bypass the charismatically-challenged intermediaries in the formal chain of command and get the message out to the next hierarchical level of followers directly, perhaps by implementing a management by walking around technique (Peters & Austin 1985). A proactive approach that top-level leaders could take to ensure that their vision is properly disseminated to followers with the proper enthusiasm and energy is to include charisma as a selection criterion when hiring or promoting intermediate-level leaders. It might behoove charismatically-challenged leaders to enlist a cadre of charismatic frontline subordinates who are fully capable of spreading the word throughout the organization so that it eventually reaches the bottom-level followers. In the movie, The Godfather (Copolla, 1972), the stoic godfather, Don Vito Corleone, successfully (although perhaps not intentionally) uses this technique to get the message across, so to speakhis henchmen were usually very effective at conveying a forceful yet compelling message to followers, regardless of the legality of the actual techniques used.


71 Another practical application of the results of this study is to provide charisma training to intermediate-level leaders. Research indicates that charismatic behaviors can be learned (Howell & Frost, 1989), suggesting that intermediate-level leaders not possessing sufficient charismatic qualities can be trained to be more charismatic, effectively enhancing the mediation process introduced in this dissertation. The expected results of such a strategy would be twofold: either the trained intermediate-level leaders would 1) augment the charismatic influence of top-level charismatic leaders, or 2) substitute for the charisma of a top-level leader who lacks charisma. This line of discussion begs a second question: do upper-level leaders need to be charismatic? Consider some leaders of contemporary corporate success stories. Few would rate Bill Gates high on the charisma scale, yet he has successfully built one of the largest organizations in the world. How was he able to successfully convey his successful vision to such a huge organization without charisma? His many innovative ideas and visions were obviously conveyed to the right people, most likely intermediate-level leaders. One could speculate that he did not do it alone, but that he relied on charismatic subordinates who were responsible for effectively spreading the messages to the followers. Limitations The most obvious limitations were due to methodology and were recognized during the initial data review. In study 1, the sample size was rather small and the ratio of the number of intermediate-level leaders to upper-level leaders was inadequate. In Study 2, as discussed in the methods section previously, each cadet squadron consists of three main flights (A, B, and C) and one smaller administrative flight consisting of cadets in staff positions within the squadron. The Internet-based survey limited responses on flight


72 membership to only A, B, or C, and not the administrative flight. It is unclear how the cadets in staff positions selected a flight, and if they then proceeded to rate the leader of the flight they indicated (i.e., A, B, or C) or the leader of the administrative flight. Nonetheless, considering the relatively small number of cadets in this situation, it is likely that the error introduced was random versus systematic, and would not be expected to bias the results in any specific direction. A more elusive limitation to this study is the fact that it is difficult to separate or distinguish the effects of proximal dyadic charismatic effects from distal charismatic effects. The influence observed on bottom-level followers is no doubt a combination of both distal and proximaldissecting the two effects is challenging. This suggests that the effect sizes observed in the extant charismatic leadership literature are also a combination of both proximal and distal dyadic effects, since the studies did not discriminate between the two effects. However, considering the arguments introduced in this dissertation, it is likely that the more proximal dyadic effects of charisma account for a larger majority of the influence on followers, especially as the size of the organization increases. As emphasized previously, top-level leaders are unable to maintain direct-contact relationships with the large majority of the followers, and are only able to influence them through brief, intermittent, and distant communications. These distal relationships cannot sufficiently explain long term influences. Instead, the more proximal and frequent contact leader-follower relationships are better suited to provide the sustained leader influences. Therefore, one would expect such proximal dyad processes to account for a majority of the influence at the follower level. Finally, if proximal dyad processes are a major source of influence at the follower level, the suggested mediation processes and associated


73 intermediaries become crucial components in the process by which charisma cascades down through and organization. Another obvious limitation is that I only investigated two mediating variables at the intermediate level and an adequate theory of leadership should include many more intermediate-level leader variables. In addition, not every intermediate-level leader will be a conduit of upper-level leader charisma to the same extent, so a comprehensive theory of leadership mediation should also include salient moderating variables. Future Research The new approach to charismatic leadership presented in this dissertation only fills a minor hole in the charismatic leadership literature. While charisma can be a complex and often obscure construct, each additional piece of information we learn about the charismatic process becomes useful in the pursuit of a comprehensive charismatic theory. This new approach raises issues that must not be neglected in future research, but also leaves many issues unresolved. To identify the full effects of the suggested mediation process of charisma, researchers must be able to distinguish the effects between the direct dyadic charismatic effects from the indirect or mediated charismatic effects. Until methods and experimental designs are developed to control for the effects of one or the other, as well as control for the effects of interactions between the two, further research will remain hindered. The use of additional statistical methods such as structural equations modeling (LISREL, Joreskog & Sorbom, 1996) may allow researchers to thresh out the direct effects of charisma from the indirect effects of charisma. Since the empirical data presented in this dissertation indicates intermediate-level leaders assume a significant role in the cascading charismatic leadership process, it opens a Pandora s Box full of moderating variables. All of the moderators previously applied


74 to traditional charismatic leadership research must now be applied to the intermediate-level leaders to determine the existence and extent of any interactions. Although the inclusion of some moderators at intermediate-levels may result in interactions similar to those observed in direct dyadic charismatic leader studies, researchers should not simply assume that the moderators and interactions previously identified can be applied wholesale at the intermediate-level leader. Researchers should carefully consider moderators and consider inclusion only when sufficient theoretical support exists. Considering that distant leader charisma requires mediation, perhaps other mediators exist other than intermediate-level leaders. One possible set of mediators may be symbols and slogans to transmit vision, which has been suggested by other researchers as the starting point charismatic leadership (Awamleh & Gardner, 1999). While distal charismatic leaders may not be able to directly contact each follower in the organization, the leader may be able to use important organizational symbols to pass on a charismatic message. For instance, the use of symbols like the silhouette of a light at the end of the tunnel, may effectively mediate top-level leader optimistic vision down to lower-level personnel. Likewise, catchy slogans such as Catch us if you can! may be able to instill the top-level leader attitude that the organization is out in front of their competitors. However, development of an effective slogan is difficult. Many have been exposed to meaningless vision statements released by insincere leaders, seeming as if the vision statement or slogan was a haphazardly assembled afterthought. Perhaps leaders should devote significant effort in the development of slogans which represent their vision, understanding that it may be the only exposure to the leader that some followers get.


75 Since it appears that charisma can be mediated through intermediate-level leaders, it is likely that charisma can be transferred through any individual that has sufficient exposure to upper-level leaders. Consequently, not only are intermediate-level leaders able to transfer upper-level charisma, but many others in the organization may as well. As other researchers have suggested, charisma is contagious (Cherulnik et al., 2001), and can be passed from person to person. A social contagion perspective using social network methods would be ideal for mapping the spread of charisma throughout an organization. Social network analyses will allow researchers to identify groups within the organization that share similar perceptions of leader charisma, and also determine the main communication paths for these groups. In some cases, groups may have direct but limited contact with upper-level leaders, thus allowing them to construct their perceptions of leader charisma on limited observed leader behavior. On the other hand, some groups may have no contact with upper-level leaders, suggesting that their perception of leader charisma is based solely on second hand information or indirect mediation processes. Furthermore, longitudinal experimental designs will allow more in-depth analyses of how leader charisma, or the perceptions thereof, travels throughout an organization. For instance, in organizations with newly hired or elected upper-level leaders, followers may already have perceptions of charisma prior to actually meeting the new leaders. Follower perceptions may change as the followers observe the new leaders behavior, or the perceptions may change as they are exposed to intermediate-level leaders. Conclusion This dissertation presented a new mediation approach to the charismatic leadership process that suggests that leader charisma can cascade down an organization via intermediate-level leaders. Two studies were developed to test a series of hierarchical


76 models. While the results of the two studies were not totally consistent, the results from Study 2 provided robust results indicating that the relationship between top-level leader charisma and bottom-level follower satisfaction was partially mediated by intermediate-level leader charisma and satisfaction. This result suggests that in addition to the direct dyadic effects of top-level leader charisma, top-level leader charisma can cascade down through an organization indirectly through intermediaries. While this paper is heavily reliant on refutation of the direct dyad assumption, the results of this research do not purport to completely discount the effects associated with direct dyadic charismatic influence. Instead, this research emphasizes the importance of considering alternative mediated charismatic influence processes, perhaps simultaneously, that must be considered in charismatic leadership research.


LIST OF REFERENCES Adlemann, P. K., & Sajonc, R. B. (1989). Facial efference and the experience of emotion. Annual Review of Pyschology, 40 249-280. Antonakis, J., & Atwater, L. (2002). Leader distance: A review and a proposed theory. Leadership Quarterly, 13 673-704. Ashcraft, M. H. (1989). Human memory and cognition Glenview, IL, Scott, Foresman and Company. Avolio, B. J., & Yammarino, F. J. (1990). Operationalizing charismatic leadership using a levels-of-analysis framework. Leadership Quarterly, 1 (3), 193-208. Awamleh, R., & Gardner, W. L. (1999). Perceptions of leader charisma and effectiveness: the effects of vision content, delivery, and organizational performance. Leadership Quarterly, 10 (3), 345-373. Bandura, A. (1969). Principles of behavior modification New York: Holt, Reinhart, & Winston. Bandura, A. (1973). Aggression: A social learning analysis Englewood Cliffs, NJ: Prentice-Hall. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51 (6), 1173-1182. Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on group behavior. Administrative Science Quarterly, 47 644-675. Bass, B. M. (1985). Leadership and performance beyond expectations New York: Free Press. Bass, B. M. (1988). Evolving perspectives on charismatic leadership. In J. A. Conger & R. N Kanungo (Eds), Charismatic leadership San Francisco, CA: Jossey-Bass, Inc. Bass, B. M., & Avolio, B. J. (2000). The multifactor leadership questionnaire (2nd ed.). Redwood City, CA: Mind Garden, Inc. 77


78 Bass, B. M., Waldman, D. A., Avolio, B. J., & Bebb, M. (1987). Transformational leadership and the falling dominoes effect. Group & Organizational Studies, 12 (1), 73-87. Bliese, P. (2000). Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 349-381). San Francisco, CA: Jossey-Bass. Bliese, P. D., Halverson, R. R., & Schriesheim, C. A. (2002). Benchmarking multilevel methods in leadership: the articles, the model, and the data set. Leadership Quarterly, 13 3-14. Brief, A. P., Butcher, A. H., & Roberson, L. (1995). Cookies, disposition, and job attitudes: The effects of positive mood-inducing events and negative affectivity on job satisfaction in a field experiment. Organizational Behavior and Human Decision Processes, 62 (1), 55-62. Bryk, A., & Raudenbush, S. W. (1992). Hierarchical linear models for social and behavioral research: Applications and data analysis methods Newbury Park, CA: Sage. Butler, J. K., Cantrell, R. S., & Flick, R. J. (1999). Transformational leadership behaviors, upward trust and satisfaction in self-managed work teams. Organization Development Journal, 17 (1), 13-28. Cacioppo, J. T., & Petty, R. E. (1983). Social psychophysiology: A source book New York: Guilford Press. Chatman, J. A. (1991). Matching people and organization: Selection and socialization in public accounting firms. Administrative Science Quarterly, 36 (3), 459-484. Chatman J. A. (1989). Improving interactional organizational research: a model of person-organization fit. Academy of Management Review, 14 333-349. Cherulnik, P.D., Donley, K.A., Wiewel, T.R., & Miller, S.R. (2001). Charisma is contagious: The effect of leaders charisma on observers affect. Journal of Applied Social Psychology, 31 (10), 2149-2159. Conger, J. A. (1989). The charismatic leader: Behind the mystique of exceptional leadership San Francisco, CA: Jossey-Bass Publishers. Conger, J. A., & Kanungo, R. N. (1987). Toward a behavioral theory of charismatic leadership in organizational settings. Academy of Management Review, 12 (4), 637-647.


79 Connelly, S., Gaddis, B., & Helton-Fauth, W. (2002). A closer look at the role of emotions in transformational and charismatic leadership. In B. J Avolio & F. J. Yammarino (Eds.), Transformational and charismatic leadership: The road ahead (pp. 255-283). Oxford: Elsevier Science Ltd. Davis, M. R. (1985). Perceptual and affective reverberation components. In A. B. Goldstein & G. Y. Michaels (Eds.), Empathy: Development, training, and consequences (pp. 62-108). Hillsdale, NJ: Erlbaum. DeGroot, T., Kiker, S. D., & Cross, T. C. (2000). A meta-analysis to review organizational outcomes related to charismatic leadership. Canadian Journal of Administrative Sciences, 17 (4), 356-371. Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95 542-575. Dubinsky, A. J., Yamarino, F. J. Jolson, M. A., & Spangler, W. D. (1995). Transformational leadership: An initial investigation in sales management. Journal of Personal Selling & Sales Management, 15 (2), 17-31. Dvir, T., & Shamir, B. (2003). Follower developmental characteristics as predicting transformational leadership: A longitudinal field study. Leadership Quarterly, 14 327-344. Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes Fort Worth, TX: Harcourt Brace Jovanovich. Erez, A., Misangyi, V., Johnson, D. E., LePine, M. A., & Halverson, K. C. (2005). Stirring the hearts and minds of followers: Charismatic leadership as the transferal of positive moods. Unpublished manuscript, University of Florida, Gainesville. Erez, A., Mitchell, T. R., Jackson, C. A., & Fanelli, A., & Judge, T. A. (2005). Anticipation and recollection are better than the experience: Effects of a rosy view on event evaluations Unpublished manuscript, University of Florida, Gainesville. Etzioni, A. (1961). A comparative analysis of complex organizations New York: The Free Press. Feldman, J. M. (1981). Beyond attribution theory: Cognitive processes in performance appraisal. Journal of Applied Psychology, 66 127-148. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7 114-140. Fried, Y., & Ferris, G. R. (1987). The validity of the job characteristics model: A review and meta-analysis. Personnel Psychology, 40 (2), 287-322. Fuller, J. B., Patterson, C. E. P., Hester, K., & Stringer, D. Y. (1996). A quantitative review of research on charismatic leadership. Psychological Reports, 78 271-287.


80 Gardner, W. L., & Avolio, B. J. (1998). The charismatic relationship: A dramaturgical perspective. Academy of Management Review, 23 (1), 32-58. Hackman, J. R., & Oldham, G. R. (1975). Development of the job diagnostic survey. Journal of Applied Psychology, 60 159-170. Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Rest of a theory. Organizational Behavior and Human Performance, 16 250-279. Hatfield, E., Cacioppo, J.T., & Rapson, R.L. (1994). Emotional contagion Paris: Cambridge University Press. Herman, J. B., & Hulin, C. L. (1972). Studying organizational attitudes from individual and organizational frames of reference. Organizational Behavior and Human Performance, 8 84-108 Hogg, M. A. (2001). A social identity theory of leadership. Personality and Social Psychology Review, 5 (3), 184-200. House, R. J. (1977). A 1976 theory of charismatic leadership. In J.G. Hunt & L.L. Larson (Eds.), Leadership: The cutting edge (pp.189-207). Carbondale: Southern Illinois University Press. House, R. J., & Aditya, R. N. (1997). The social scientific study of leadership: Quo vadis? Journal of Management, 23 409-473. Howell, J. M., & Frost, P. J. (1989). A laboratory study of charismatic leadership. Organizational Behavior and Human Decision Processes, 43 243-269. Hox, J. (2002). Multilevel analysis: Techniques and applications Mahwah, NJ: Lawrence Erlbaum. Isen, A. M. (2000). Part V: Cognitive factors. In M. Lewis & J. M. Haviland-Jones. (Eds.), Handbook of emotions (pp. 417-435). New York: Guilford Press. Isen, A. M., Niedenthal, P. M., & Cantor, N. (1992). An influence of positive affect on social categorization. Motivation and Emotion, 16 (1), 65-78. Isen, A. M., Shalker, T. Clark, M., & Karp, L. (1978). Affect, accessibility of material in memory and behavior: A cognitive loop? Journal of Personality and Social Psychology, 36 1-12. Jacobson, C., & House, R. J. (2001). Dynamics of charismatic leadership: A process theory, simulation model, and tests. Leadership Quarterly, 12 75-112. Jreskog K. & Srbom, D. (1996). LISREL 8: Users reference guide Chicago: Scientific Software International.


81 Judge, T. A., & Hulin, C. L. (1993). Job satisfaction as a reflection of disposition: A multiple source causal analysis. Organizational Behavior and Human Decision Processes, 56 388-421. Judge, T. A., & Ilies, R. (2004). Affect and job satisfaction: A study of their relationship at work and at home. Journal of Applied Psychology, 89 (4) 661-673. Judge, T. A., & Larsen, R. J. (2001). Dispositional affect and job satisfaction: A review and theoretical extension. Organizational Behavior and Human Decision Processes, 86 1, 67-98. Judge, T. A., & Locke, E. A. (1993). Effect of dysfunctional thought processes on subjective well-being and job satisfaction. Journal of Applied Psychology, 78 (3), 475-490. Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 89 (5), 755-768. Kark, R., Shamir, B., & Chen, G. (2003). The two faces of transformational leadership: Empowerment and dependency. Journal of Applied Psychology, 88 246-255. Katz, D., & Kahn, R. L. (1978). The social psychology of organizations New York: Wiley. Katzell, R. A., & Thompson, D. E. (1990). Work motivation. American Psychologist, 45 144-153. Kirkpatrick, S. A., & Locke, E. A. (1996). Direct and indirect effects of three core charismatic leadership components on performance and attitudes. Journal of Applied Psychology, 81 (1), 36-51. Klein, K. J., & House, R.J. (1995). On fire: Charismatic leadership and levels of analysis. The Leadership Quarterly, 6 (2), 183-198. Komaki, J., Barwick, K. D., & Scott, L. R. (1978). A behavioral approach to occupational safety: Pinpointing and reinforcing safe performance in a food manufacturing plant. Journal of Applied Psychology, 63 (4), 434-445. Kraiger, K., Billings, R. S., & Isen, A. M. (1989). The influence of positive affective states on task perceptions and satisfaction. Organizational Behavior and Human Decision Processes, 44 12-25. Kristof, A. L. (1996). Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49 (1), 1-49.


82 Kunin, T. (1955). The construction of a new type of attitude measure. Personnel Psychology, 8 65-78. Laird, J. D. (1984). The real role of facial response in the experience of emotion: A reply to Tourangeau and Ellsworth, and others. Journal of Personality and Social Psychology, 47 909-917. LePine, J. A., & Van Dyne, L. (1998). Predicting voice behavior in work groups. Journal of Applied Psychology, 83 (6), 853-868. Lewis, M., & Haviland-Jones, J. M. (Eds.) (2000). Handbook of emotions New York, NY: Guilford Press. Locke, E. A. (1976). The nature and causes of job satisfaction. In M.D. Dunnette (Ed.), Handbook of I.O. psychology Chicago: Rand McNally. Lord, R. G., & Emrich, C. G. (2001). Thinking outside the box by looking inside the box: Extending the cognitive revolution in leadership research. Leadership Quarterly, 11 (4), 551-579. Lord, R. G., Foti, R. J., & De Vader, C. L. (1984). A test of leadership categorization theory: internal structure, information processing, and leadership perceptions. Organizational Behavior and Human Performance, 34 343-378. Lord, R. G., Foti, R. J., & Phillips, J. S. (1982). A theory of leadership categorization. In J. G. Hunt, U. Secaran, & C. Schriescheim (Eds.), Leadership: Beyond establishment views Carbondale, IL: Southern Illinois University Press. Lowe, K.B., Kroeck, K.G., & Sivasubramaniam, N.J., (1996). Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ literature. Leadership Quarterly, 7 (3), 385-425. Madsen, D., & Snow, P. G. (1983). The dispersion of charisma. Comparative Political Studies, 16 (3), 333-362. McHugo, G. J., Lanzetta, J. T., Sullivan, D. G., Masters, R. D., & Englis, B. G. (1985). Emotional reactions to a political leaders expressive displays. Journal of Personality and Social Psychology, 49 1513-1529. Meindl, J. R. (1990). On leadership: An alternative to the conventional wisdom. Research in Organizational Behavior, 12 159-203. Muchinsky, P. M. & Monahan, C. J. (1987). What is person-environment congruence? Journal of Vocational Behavior, 31 268-277. Murphy, K. R., Jako, R. A., & Anhalt, R. L. (1993). Nature and consequences of halo error: A critical analysis. Journal of Applied Psychology, 78 (2), 218-225.


83 Newcomb, T. M. (1943). Personality and social change: Attitude formation in a student community New York: Dryden. Niehoff, B. P., Enz, C. A., & Grover, R. A. (1990). The impact of top-management actions on employee attitudes and perceptions. Group & Organization Studies, 15 (3), 337-352. OReilly, C. A., Chatman, J., & Caldwell, D. F. (1991). People and organizational culture: A profile comparison approach to assessing person-organization fit. Academy of Management Journal, 34 (3), 487-516. OReilly, C. A. & Roberts, K. H. (1975). Individual differences in personality, position in the organization, and job satisfaction. Organizational Behavior and Human Performance, 14 144-150. Percy, P. M. (1997). Relationships between interpersonal power and followers satisfaction: A leadership perspective. Dissertation Abstracts International A, 58(6-A). Peters, T., & Austin, N. (1985). Management by walking around. California Management Review, 28 (1), 9-34. Pillai, R., Schriesheim, C. A., & Williams, E. S. (1999). Fairness perceptions and trust as mediators for transformational and transactional leadership: A two-sample study. Journal of Management, 25 (6), 897-933. Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader behaviors and their effects on followers trust in leader, satisfaction, and organizational citizenship behavior. Leadership Quarterly, 1 (2), 107-142. Podsakoff, P. M., MacKenzie, S. B., & Bommer, W. H. (1996). Meta-analysis of the relationship between Kerr and Jermiers substitutes for leadership and employee job attitudes, role perceptions, and performance. Journal of Applied Psychology, 81 (4), 380-399. Provine, R. R. (1986). Yawning as a stereotyped action pattern and releasing stimulus. Ethology, 72 109-122. Provine, R. R. (1992). Contagious laughter: Laughter is a sufficient stimulus for laughs and smiles. Bulletin for the Psychonomic Society, 30 1-4. Puzo, M, (Writer), & Coppola, F. F. (Director). (1972). The Godfather [Film]. (Available from Paramount Pictures, Hollywood, CA) Ross, S. M., & Offermann, L. R. (1997). Transformational leaders: Measurement of personality attributes and work group performance. Personality & Social Psychology Bulletin, 23 (10), 1078-1086.


84 Rosch, E. (1973). On the internal structure of perceptual and semantic categories. In T. E. Moore (Ed.), Cognitive development and the acquisition of language New York, NY: Academic Press. Rosch, E. H. (1975). Cognitive representations of semantic categories. Journal of Experimental Psychology: General, 104 192-233. Salancik, G. R. & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 23 224-253. Schein, E. H. (1990). Organizational culture and leadership San Francisco: Josey-Bass Publishers. Schriesheim, C. A., Castro, S. L., Zhou, X., & Yammarino, F. J. (2001). The folly of theorizing A but testing B: A selective level-of-analysis review of the field and a detailed leader-member exchange illustration. Leadership Quarterly, 12 515-551. Shamir, B., House, R. J., & Arthur, M. B. (1993). The motivational effects of charismatic leadership: A self-concept based theory. Organization Science, 4 (4), 577-594. Shamir, B. (1995). Social distance and charisma: Theoretical notes and an exploratory study. Leadership Quarterly, 6 (1), 19-47. Shamir, B., Zakay, E., Breinin, E., & Popper, M. (1998). Correlates of charismatic leader behavior in military units: Subordinates attitudes, unit characteristics, and superiors appraisals of leader performance. Academy of Management Journal, 41 (4), 387-409. Smith, R. C., Kendall, L. M., & Hulin, C. L. (1969). The measurement of satisfaction in work and retirement Chicago, IL: Rand McNally & Company. Staw, B. M., Bell, N. E., & Clausen, J. A. (1986). The dispositional approach to job attitudes: A lifetime longitudinal test. Administrative Science Quarterly, 31 56-77. Staw, B. M., & Ross, J. (1985). Stability in the midst of change: A dispositional approach to job attitudes. Journal of Applied Psychology, 70 (3), 469-480. Stotland, E. (1969). Exploratory investigations of empathy. In L. Berkowitz (Ed.), Advances in experimental social psychology (pp. 271-314). New York, NY: Academic Press. Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 681-706). Hillsdale, NJ: Erlbaum. Thorndike, E. L. (1920). A constant error in psychological ratings. Journal of Applied Psychology, 4 25-29.


85 Totterdell, P., Kellett, S., Teuchmann, K., & Briner, R. B. (1998). Evidence of mood linkage in work groups. Journal of Personality and Social Psychology, 74 (6), 1504-1515. Waldman, D. A., & Yammarino, F. J. (1999). CEO charismatic leadership: Levels-of-analysis management and levels-of-analysis effects. Academy of Management Review, 24 (2), 266-285. Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience aversive emotional states. Psychological Bulletin, 96(3), 465-490. Watson, D., & Slack, A. K. (1993). General factors of affective temperament and their relation to job satisfaction over time. Organizational Behavior and Human Decision Processes, 54 181-202. Weber, M. (1947). The theory of social and economic organization Trans. by A. M. Henderson & T. Parsons. New York, NY: The Free Press. Weick, K. E. (1977). Enactment processes in organizations. In B. M. Staw & G. R. Salancik (Eds.), New directions in organizational behavior Chicago, IL: St. Clair Press. Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion of the structure, causes and consequences of affective experiences at work. Research in Organizational Behavior, 18 1-74. Weiss, H. M., Nicholas, J. P., & Daus, C. S. (1999). An examination of the joint effects of affective experiences and job beliefs on job satisfaction and variations in affective experiences over time. Organizational Behavior & Human Decision Processes, 78 (1), 1-24. Willner, A. R. (1968). Charismatic political leadership: A theory Princeton, NJ: Princeton University, Center for International Studies. Yammarino, F. J. (1994). Indirect leadership: Transformational leadership at a distance. In B. M. Bass & B. J. Avolio (Eds.), Improving organizational effectiveness through transformational leadership (pp. 26-47). Thousand Oaks, CA: Sage. Yukl, G. (2002). Leadership in organizations (5th ed.) Upper Saddle River, NJ: Prentice-Hall.


BIOGRAPHICAL SKETCH Kent C. Halverson was born in Edina, Minnesota, and raised in Northfield, Minnesota, where he graduated from Northfield High School in 1986. He entered the Air Force and received a Bachelor of Science degree in civil engineering and was commissioned as an Air Force officer from the United States Air Force Academy in Colorado Springs, Colorado, in 1990. He received a Master of Science degree in structural engineering from the University of Illinois at Champaign-Urbana, Illinois, in 1995. Throughout his military career he has served in various civil engineering capacities at Hansom Air Base in Bedford, Massachusetts, Ramstein Air Base, Germany, Cape Canaveral Air Station, Florida, Kunsan Air Base, Republic of South Korea, and the Air Force Institute of Technology, Dayton, Ohio. He earned his PhD from the Warrington College of Business at the University of Florida in August 2005, and is currently on the faculty at the Air Force Institute of Technology in Dayton, Ohio. 86