<%BANNER%>

Psychometric Properties of the Norbeck Social Support Questionnaire

Permanent Link: http://ufdc.ufl.edu/UFE0022029/00001

Material Information

Title: Psychometric Properties of the Norbeck Social Support Questionnaire
Physical Description: 1 online resource (39 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: adhd, psychometrics, reliability, social, validity
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Social support is a frequently measured construct though many existing measures of the construct are lacking in psychometric support. The Norbeck Social Support Questionnaire (NSSQ) is a promising, multidimensional measure of social support that has been shown to be highly reliable with moderate validity in nursing students. Perceived support of mothers with children at high risk for or with ADHD diagnoses was assessed using the NSSQ at three time points (Time 1 n = 266, Time 2 n = 207, Time 3 n = 49). The measure demonstrated high internal consistency but low stability across two- and four-year intervals. Convergent and predictive validity of the NSSQ was examined using measures of caregiver strain and general mental well-being. Analyses produced weak relationships, suggesting that while the NSSQ demonstrates internal reliability, it does not adequately relate to the theoretically linked constructs of strain or general psychological well-being. These findings may reflect a need to include information about network characteristics in psychometric analyses. Future revisions may also be needed to include other relevant types of support.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Eyberg, Sheila M.
Local: Co-adviser: Bussing, Regina.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022029:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022029/00001

Material Information

Title: Psychometric Properties of the Norbeck Social Support Questionnaire
Physical Description: 1 online resource (39 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: adhd, psychometrics, reliability, social, validity
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Social support is a frequently measured construct though many existing measures of the construct are lacking in psychometric support. The Norbeck Social Support Questionnaire (NSSQ) is a promising, multidimensional measure of social support that has been shown to be highly reliable with moderate validity in nursing students. Perceived support of mothers with children at high risk for or with ADHD diagnoses was assessed using the NSSQ at three time points (Time 1 n = 266, Time 2 n = 207, Time 3 n = 49). The measure demonstrated high internal consistency but low stability across two- and four-year intervals. Convergent and predictive validity of the NSSQ was examined using measures of caregiver strain and general mental well-being. Analyses produced weak relationships, suggesting that while the NSSQ demonstrates internal reliability, it does not adequately relate to the theoretically linked constructs of strain or general psychological well-being. These findings may reflect a need to include information about network characteristics in psychometric analyses. Future revisions may also be needed to include other relevant types of support.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Eyberg, Sheila M.
Local: Co-adviser: Bussing, Regina.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022029:00001


This item has the following downloads:


Full Text

PAGE 1

PSYCHOMETRIC PROPERTIES OF THE NORBECK SOCIAL SUPPORT QUESTIONNAIRE By MONICA LEAH STEVENS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Monica Leah Stevens 2

PAGE 3

ACKNOWLEDGMENTS I thank Regina Bussing and Sheila Eyberg fo r their mentorship, editorial support, and encouragement as I worked on this thesis. I would also like to thank James Johnson, Michael Perri, and Brenda Weins, members of my ma sters thesis committee, for their thoughtful feedback, time, and support. I owe many thanks to Cynthia Garvan for he r generous statistical guidance and expertise. Finally, I want to acknowl edge the National Institute of Mental Health (RO1 MH57399) for funding the project from which data for this thesis were drawn. 3

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........6 ABSTRACT.....................................................................................................................................7 CHAPTER 1 INTRODUCTION................................................................................................................. ...8 The Impact of ADHD on Parenting..........................................................................................9 Parental Social Support and Child Behavior Problems............................................................9 Issues in the Measurement of Social Support.........................................................................10 Necessary Components of Support Measurement..................................................................12 Perceived versus Received Social Support.............................................................................13 The Norbeck Social Support Questionnaire...........................................................................13 Specific Aims..........................................................................................................................17 2 METHODS...................................................................................................................... .......18 Participants.............................................................................................................................18 Measures.................................................................................................................................19 Procedure................................................................................................................................21 Data Analysis..........................................................................................................................22 Data Integrity...................................................................................................................22 Reliability Analysis.........................................................................................................22 Validity Analysis.............................................................................................................2 2 Convergent validity..................................................................................................22 Predictive validity....................................................................................................23 3 RESULTS...................................................................................................................... .........24 Reliability...............................................................................................................................24 Convergent Validity............................................................................................................ ....25 Predictive Validity............................................................................................................ ......25 4 DISCUSSION................................................................................................................... ......27 APPENDIX A NSSQ ITEMS................................................................................................................... ......31 B ADDITIONAL DATA...........................................................................................................32 4

PAGE 5

LIST OF REFERENCES...............................................................................................................35 BIOGRAPHICAL SKETCH.........................................................................................................39 5

PAGE 6

LIST OF TABLES Table Page A-1 NSSQ Items used to Rate the Quality of Social Support Provided by Members of the Respondents Social Support Network..............................................................................31B-1 Time 1 and Time 2 Means, Standard Deviations, Ranges of Scores.................................32B-2 Pearson Correlation Coefficients for all Pairs of Items on Time1 NSSQ..........................32B-3 Test-Retest Stability Pears on Correlations for NSSQ Scales............................................33B-4 Time 1 Pearson Correlations Between the NSSQ Scales and Specific Caregiver Strain Questionnaire Subscales ( n = 262)..........................................................................33B-5 Time 2 Pearson Correlations Between th e NSSQ Scales and the Mental Health Inventory-5 Total Score ( n = 200).....................................................................................33B-6 NSSQ Scales Predictive Valid ity of Specific CSQ Subscales ( n = 138)...........................34B-7 NSSQ Scales Predictive Validity of Me ntal Health Inventory-5 Total Scores ( n = 114)....................................................................................................................................34 6

PAGE 7

7 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science PSYCHOMETRIC PROPERTIES OF THE NORBECK SOCIAL SUPPORT QUESTIONNAIRE By Monica Leah Stevens May 2008 Chair: Regina Bussing Cochair: Sheila Eyberg Major: Psychology Social support is a frequently measured construct though many existing measures of the construct are lacking in psychometric support. The Norbeck Social Support Questionnaire (NSSQ) is a promising, multidimensional measure of social support that has been shown to be highly reliable with moderate validity in nursi ng students. Perceived s upport of mothers with children at high risk for or with ADHD diagnoses was assessed using the NSSQ at three time points (Time 1 n = 266, Time 2 n = 207, Time 3 n = 49). The measure demonstrated high internal consistency but low stability across twoand f our-year intervals. Conve rgent and predictive validity of the NSSQ was examined using measures of caregiver strain and general mental wellbeing. Analyses produced weak relationships, su ggesting that while th e NSSQ demonstrates internal reliability, it does not adequately relate to the theoretically linked c onstructs of strain or general psychological well-being. These findings may reflect a need to include information about network characteristics in psychometric analyses. Fu ture revisions may also be needed to include other relevant types of support.

PAGE 8

CHAPTER 1 INTRODUCTION Although definitions of social support vary, comm on to most definitions is the exchange of assistance or nurturance and emotional support be tween two or more indi viduals (Tietjen, 1985). Social support is related to better health ou tcomes across a multitude of conditions (e.g., Miller, McMahon, &Garrett, 1989; Rice & Longabaugh, 1996; Dormire, Strauss, & Clarke, 1989), including parent and child mental health (Hanson, 1986). Th e literature includes findings on social support and parental outcomes such as maternal depression (Cairney, Boyle, Offord, & Racine, 2003), as well as general child emotiona l and social outcomes (Devall, Stoneman, & Brody, 1986), and reveals a strong li nk between psychological well-being and satisfaction with perceived support systems. Within families, there are clear connections between parents social support systems and parenting stress (F ischer, 1990; Bussing, et al., 2003). Despite the growing body of res earch on social support, researchers have yet to establish reliable and valid instruments to measure the co nstruct. A barrier to th e creation of adequate measures is the variability not only in the defi nition of support but in th e function of support for different populations. One such population is parents of children with attention deficit hyperactivity disorder, a type of externalizing diso rder associated with in creased activity levels, inattention, and impulsivity (4th ed.; DSM-IV; American Psychiatric Association, 1994). These parents are known to experience hi gh levels of parenting stress, engage in less positive parenting techniques, and have a lower quali ty of life than parents of ch ildren without clinical problems (Fischer, 1990; Lange, 2005). Because of these fi ndings, further research is needed on factors that contribute to positive paren ting experiences in this population. 8

PAGE 9

The Impact of ADHD on Parenting Attention deficit hyperactivity diso rder is a chronic disorder th at occurs in three to five percent of school-aged children. The disorder is characterized by developm entally inappropriate levels of inattention, impulsivity, and hyperactiv ity. The features of ADH D often leave children at risk for academic problems, interpersonal diffi culties, and significant be havioral issues that compromise their quality of lif e (Treacy, Tripp, & Baird, 2005). The characteristics of ADHD not only affect th e child but have been linked to increased familial difficulties. Parents of children with ADHD tend to report less parenting satisfaction, display more ineffective discipline strategies, and have more marital conflict than parents of children without externalizing behavior disorders (Lange, et al, 2005; Suarez & Baker, 1997). Not surprisingly, these parents are more vulnerable to elevated levels of stress compared with parents of normal controls (Eyberg, Boggs, & Rodriguez, 1992; Treacy, Tripp, & Baird, 2005). Research has shown higher levels of parenting st ress to be associated with negative parenting practices including physical punishment (Lange et al, 2005; McCurdy, 2005), which puts into effect a maladaptive cycle between child behavior and parenting approaches. Parental Social Support and Child Behavior Problems Not only are parents of children with problem be haviors typically more stressed, they tend to be more socially isolated. Children with ADHD of ten demand more parental resources than other children, which can lead parents to neglect other areas of their ow n life. In fact, Lange et al. (2005) found that parents of children with ADHD have less total social suppor t, particularly from family and friends, more stress, and a lower quality of life than other parents. Inadequate social support compromises a parents ability to monitor, classify, and respond to child behavior (Wahler, 1980; Wahler and Dumas, 1989). Colletta (1979) reports that mothers with low support tend to be more restrictive and punitive in th eir parenting. According to the stress-buffering 9

PAGE 10

hypothesis, however, positively perceived social ne tworks have been found to buffer stress and promote competent parenting behavior (Rodgers, 1998). Issues in the Measurement of Social Support There is considerable variab ility in thought about how best to measure social support (Langford, Bowsher, Maloney, & Lillis, 1997). Lite rature on measurement of support in parents in particular is limited but nevertheless suggest s that instruments designed to measure support should be developed to reflect the relevant characteristics of a parents support system. Identifying those relevant characteristics is an increasingly complex task when families are faced with the challenges of child pr oblem behaviors. Because support measures designed specifically for parents are rare, consideration has to be given to existing studies of parental social support in order to identify specific facets of pare ntal support that should be measured. Efforts to address the importance of parental social support in treatment outcome research have had mixed success. Dadds and McHugh (1992) found that parents with higher levels of social support from friends generally showed greater improvement in pa rent-training. However, in this same study they found that the addition of a social support enhancement component to treatment was not effective in improving treatment outcomes. Considering the complex nature of social support, these latter findings suggest there may be a brea kdown in understanding of social support, as the support measure used was a global index of support rather than a more narrowlydefined, multidimensional tool. Because social support is a complex conc ept, and sources of support most beneficial to indi viduals vary, measurement of support should be comprehensive enough to inform treatment plans effectively. Despite this, many instruments designed to measure support do not reflect this multidimensional model of support (Winemiller, Mitchell, Sutliff, and Cline, 1993). 10

PAGE 11

Part of this complexity exists because va rious types of social support affect families differently depending on demographic characteristics of the famil y. Wan, Jaccard, and Ramey (1996) found that married mothers report benefiti ng more from familial sources of support while single mothers and fathers relied more on non-familial support. Suarez and Baker (1997) examined the effects of parental support on marri ed parents perception of parenting and their perception of their childs impact on the family life. In their study, spousal support predicted how parents perceived their role as parents and how they percei ved the impact of their childs behavior on day-to-day activity in that less spousal support resu lted in more negative views on both outcome measures. In this study, global so cial support (i.e., including support outside the marriage) was not related to the outcome measures sugge sting that spousal s upport is of primary importance when available, and that parents w ithout spousal support rely on other sources. Differences exist not only in different family st ructures but across ethnic ities. Bussing, et al. (2003) compared support network characterist ics across African American and Caucasian families with ADHD children and whether these char acteristics affect health-seeking behaviors. This study found significant differences between racial groups. African American families tended to have smaller support networks than Ca ucasian families, and within these networks, they reported less professional support sources (e.g., counselors, physicians). However, they reported more frequent contact with support network members than Caucasian families and higher levels of emotional and instrumental support, meaning they reported feeling more appreciated and supported as well as having more access to tangible support than their Caucasian counterparts (Bussing, et al., 2003). The findings from both Wa n et al. and Bussing et al. highlight the need for comprehensive measuremen t of social support that addresses the multiple domains of support important within diverse familie s. A single instrument capable of capturing 11

PAGE 12

family social support from these multiple domai ns would provide more useful and accurate information regarding parental support networks. Necessary Components of Support Measurement Researchers in the area of social support tend to agree that support is a multidimensional concept, but there is a wide range of accepted definitions for the construct (Sarason, Shearin, Pierce, & Sarason, 1987; Orth-Gomer & Unden, 1987). These definitional differences lead to a variety of formats for the measurement of support, which may hinder comparability across studies. There are, however, frequently measured types of support, which commonly include (a) tangible or aid-related support, (b) informational support, (c) em otional support, and (d) social companionship (Orth-Gomer & Unden, 1987; Winemiller, Mitchell, Sutliff, & Cline, 1993). Though definitions of support vary, the complex nature of support is well-acknowledged. An instrument capable of distinguishing the composition of support systems by type of support provided and the specific sour ces of support (e.g., tangible su pport or health care provider support) may be more useful that a global index of support alone (Winemiller, Mitchell, Sutliff, & Cline, 1993). In addition to the need for instruments w ith more sophisticated operationalization of social support, there is also a need for instruments with adequate psychometric support. One concern within the social support literature ha s been the widespread use of measures with inadequately established psychometric prope rties (Norbeck, Linds ey, & Carrieri, 1981; Winemiller, Mitchell, Sutliff, & Cline, 1993). Also, social support instruments should be standardized within multiple populations, particular ly at-risk populations, to increase the utility of these measures in clinical settings (Norbeck, Lindsey, & Carri eri, 1982). Finally, reliable and valid measures of support are needed to further understanding of social support within clinical research (Norbeck, Lindsey, & Carrieri, 1981). 12

PAGE 13

Perceived versus Received Social Support Important distinctions in perc eived and received support have been identified in response to debate over which type, perceived or receive d, of support is most rele vant. These constructs consistently emerge as unique fr om one another showing moderate correlations to each other at best (Lakey, et al, 2002). Percei ved social support is generally thought to be a more clinically relevant concept than received support. Cohe n and Hoberman (1983) find perceived support to be more related to outcome variables such as life stress. In addition, Sandler and Barrera (1984) found perceived support to better predict psychop athology symptomatology than scales assessing actual received support and support network size Received support has been found less effective as a buffer against the adverse impact of lif e stress on psychological well-being (Coventry, Gillespie, Heath, & Martin, 2004). The Norbeck Social Support Questionnaire The Norbeck Social Support Questionnaire (NSSQ) is a comprehensive, theoretically grounded instrument for measuring social suppor t with promising psychometric findings. The NSSQ is based on Kahns (1979) definition of suppor t as interpersonal transactions that include one or more of the following: the expression of positive affect of one person toward another; the affirmation or endorsement of another persons behavior, perceptions, or expressed views; the giving of symbolic or material aid to another (85). Kahn descri bed support provision as being delivered through a persons convoy, a metaphorical description of a su pport network member. Therefore, Kahns conceptualiza tion of support includes not only pos itive regard and expressions of approval but also of the provi sion of tangible aid. Kahns con ceptualization also included any individuals in a persons life that they rely on for emotional a nd tangible support, and who rely on them in the same way, so that support is seen as a reciprocal relationship (Norbeck, Lindsey, & Carrieri, 1981). 13

PAGE 14

With Kahns multidimensional support model as a guideline, Norbeck et al. developed the NSSQ to assess an individuals perceived emotional and tangible s upport. The NSSQ was also designed to gather informa tion about the size of a persons social network, the frequency of contact with network members, and network memb er loss. To complete the NSSQ, participants are presented with a series of half-pages that are aligned with a partic ipants personal network list. This format was intended to simplify the task of recalling each member of ones support network and subsequently answering a series of questions about that person. Respondents are first asked to list up to 20 signifi cant people in their life who they currently consider important to them on the right side of the page using only init ials or first names. There is also space for identification of the category in which that person falls (e.g. family, neighbors, spiritual leader, or health care provider) using a provided list of categories. Participants are next asked to rate each network member using a 5-point Likert scale (with zero indicating not at all and four indica ting a great deal). Two questions were developed for measuring each of the three functional properties of social support: (a) affect, (b) aid, and (c) affirmation. For example, one of the two affect questions asks How much does this person make you feel liked or loved? Ratings on each pair of questions are summed to yield subscale total scores (i.e., Aid total and Emotional Support total, which comprises affect and affirmation). To assess the size, stability, and availability of the support ne twork, it is possible to count the number of network members listed and examine frequency and duration of network member contact through self-repo rted numerical ratings. To further explore the network dynamics, respondents indicate recent losses of important relationships by reporting number of lost support members as well as the type of suppor t lost (i.e. aid, affective, or affirmative) and 14

PAGE 15

the amount of support that member provides using a 5-point Likert rating scale. The NSSQ can be easily scored using th e response sheet provided. The instrument was initially standardized us ing 75 first-year gra duate nursing students including one male, with a mean age of 30.3 and a range of 23-51 years. Participants were Caucasian (92%), Asian (3%), and Hispanic (1%) in ethnicity. Most were single and never married (43%), with the remaining participants representing married (3 7%) and divorced or separated (20%) adults. This gr oup was first tested after one week of enrollment in the nursing program, presumably before peer relationships were formed. For comparison, a group of 60 senior-year nursing studen ts including six males, with a m ean age of 27.3 and a range of 21-39 years, was also given the NSSQ. This groups et hnicities were Caucasian (80%), Asian (17%), and other (3%). Most participants in this gr oup were also single and never married (63%) and the remaining were married (27%) and divorced or separated (10%). Items on the NSSQ were collapsed into the th ree variables: Total F unctional (the sum of each Aid, Affect, and Affirmation items), To tal Network (number in network, length of relationship, and frequency of contact with network member), and Total Loss (number of categories of persons lost and amount of support lo st). To evaluate test -retest reliability, the NSSQ was re-administered one w eek later, and test-r etest reliability was high, ranging from .85.92. Internal consistency was eval uated through inter-item correlati ons. Item pair correlations ranged from .72 to .98 with aid being the least correlated to other subscales. Affect and Affirmation were found to be highly correlated (r ange: .95 to .98), suggesting these might not be distinct variables. Due to consis tent findings that these subscales are highly related, an Emotional Support subscale is now calculated by combin ing these items. Network property items (e.g., duration of relationships) were highly related (.88 to .97) to Affect and Affirmation and 15

PAGE 16

moderately related to aid (.69 to .80). Loss items (e.g., quantity and quality of network member loss) were not significantly related to support types or network property items (.54 to .68) demonstrating that while network composition fluc tuates, type of support is relatively stable. Moderate concurrent validity was found when the NSSQ was administered with the Social Support Questionnaire (SSQ; Shaefer, Coyne, & Lazarus, 1981). Correlations range from weak to significantly moderate relationships wh en the measure was administered along with the SSQ. Construct validity was not adequately esta blished when the NSSQ was compared to the Profile of Mood States (POMS; McNair, Lorr, & Droppleman, 1971) and the Life Experiences Survey (LES; Sarason, Johnson, & Siegel, 1978). So cial desirability bias was examined through analysis of concurrent completion of the Marlowe-Crown Test of Social Desirability Scale (Crowne & Marlowe, 1960) and it was found that none of the NSSQ items correlated significantly with this measure (N orbeck, Lindsey, & Carrieri, 1981). The NSSQ was further standa rdized within additional popu lations includ ing university staff members (Norbeck, Lindsey, & Carrieri, 1982) and mothers enrolled in college (Gigliotti, 2002, 2006). Findings indicated the need for furt her specification of highly endorsed support types such as friends and spouses. For instance, 61% of respondents listed a support member as belonging in the category of s pouse, although only 42% identified themselves as married, suggesting that additional categories represen ting significant others may be useful. The investigators also found that 90% of participants did not list a health car e provider as a network support member, a finding that has implications for future efforts at understanding the support role of health care providers. These studies confirmed earlier findings of good test-retest reliability and went on to establish concurrent and predictive validity. Moderate to high seven-month test-retest reliability 16

PAGE 17

17 was established indicating stabil ity over extended periods of time. Concurrent validity was moderately demonstrated using the Personal Re source Questionnaire (PRQ; Brandt and Weinert, 1981). Correlations between the NSSQ variables a nd the total score on the PRQ were calculated and represented medium levels of association and ranged from .35 to .41 (Norbeck, Lindsey, & Carrieri, 1982). Some evidence of predictive va lidity was demonstrated by aid-related support accounting for 13.2% of the variance when pred icting life stress (LES; Sarason, Johnson, & Siegel, 1978). Initial exploratory factor analys es revealed that the proposed functional support types (i.e., aid, affect, and affirmative) re present unique dimensions important to the understanding of social support networks (Gig liotti, 2002). Subsequent factor analyses suggested, however, that only Aid measures uniq ue dimensions of support (Gigliotti, 2006), and the support subtype, Emotional Support, was created to represent reports of affirmational and affective support. Specific Aims The purpose of this study was to examine the reliability as we ll as convergent and predictive validity of the Norbeck Social Support Questionnaire utilizing responses from mothers of children with or at high risk for ADHD. Because parents of children with ADHD represent a population with unique characteristics, it is importa nt to study the psychometric properties of the NSSQ as a measure of support in this group across time. Internal consistency and stability of the measure at intervals up to twoand four-years we re calculated as measures of test reliability. Because social support is thought to be related to stress, convergent as well as predictive validity were tested using a measure of caregiver strai n. Additionally, convergent and predictive validity were examined using the related construc t of overall parental mental health.

PAGE 18

CHAPTER 2 METHODS Participants Participants for this study were drawn from a large longitudinal study examining factors related to help-seeking behaviors and treatment continuity in children and adolescents with ADHD (Bussing et al., 2003). To be eligible for participation in the longitudinal study, families had to have a telephone in their home, be Caucas ian or African American, and only one child in the home could participate in the study. Families we re excluded from partic ipation if the target child was receiving special educatio n services for mental retardation or autism. Participants were parents of children at high risk for ADHD. Initially, a stratified random sample of 1615 elementary school children was recruited, with girls over sampled by a ratio of two to one to account for lower ADHD prevalence rates among girls. From this sample, 266, 207, and 49 mothers of high risk children have completed the NSSQ at Time 1, two year s later at Time 2, and four years later at Time 3, respectively. High risk for ADHD was defined by the presence of one or more of the following: (a) parent report of an existing ADHD diagnosis or on-going professional treatment for ADHD; (b) a suspicion of ADHD by family or school officials or (c) parent or teacher ratings on behavior problem subscales of the SNAP, a measure of ADHD symptomatology, at least 2 SD above the norm for the childs age and gender. Diagnoses upon enrollment were determined at Time 1 through parent interview using the Diagnostic Interview Schedule for Children, Version 4.0 (DISC-4). At Time 1, 62% of children met DSM-IV criteria for ADHD. Comorbid diagnoses within this sample included conduct disorder (15%) and oppositional defiant disorder (60%). At Time 1 of the longitudina l study, all children we re enrolled in grades K-5 at schools within 30 miles of Gainesville, Florida. Among th e high risk children identified at Time 1, 55% 18

PAGE 19

were female and 63% were Caucasian, with the remaining 36% African American. Socioeconomic status scores were calculated using th e Hollingshead 4-factor index on which scores can range from 8 for the lowest social strata to 66 for the highest social strata (Hollingshead, 1975). In this sample, 42% of families were classi fied as low SES as defined by a score of 33 or below. Demographic characterist ics of the Time 1 sample ar e representative of the 2000-2001 U.S. Census Data for families in Alachua County, Florida, which is similar to data for Florida as a whole and generally representative of the United States in general. Measures The Norbeck Social Support Questionnaire (NSSQ, Norbeck, Lindsey, & Carrieri, 1981) was developed as a self-administered measure of perceived emotional and tangible support. The measure was adapted for the longitudinal study by adding parent-specific words or phrases to four of the items, to determine parents percei ved level of network support in reference to their caregiver role. In this adapta tion, support network was define d as people available for the caregiver to talk with and re ly on particularly when concer ned about their childs health, behavior, or emotions. The four items were alte red to target parents rather than a generic population through these modifications. See Table A1 in Appendix A for the complete listing of functional support and network items. Adapta tions are indicated in bold text. The original NSSQ was devel oped in self-administration form. However, in this study, research assistants administered the NSSQ in in terview format. Parents were first asked to list up to 12 significant people in their life whom they currently considered important to them in their caregiver role. They were then asked to rate each of these people on 6 items, which make up the NSSQ functional subscale s, using an answer card displaying the five response categories (0 = not at all, 1 = a little 2 = moderately 3 = quite a bit 4 = a great deal) The items pertain to amount of perceived support provided by that person. Three additional items are used to assess length of 19

PAGE 20

relationship, contact frequency, and geographical distance from that person (see items 7 through 10 in Table A1). NSSQ completion time varied according to the number of support network members listed, with a maximum administ ration time of approximately 20 minutes. Three types of support are measured by NSSQ functional subscales : Affective (feeling liked or loved), Affirmational (ability to conf ide and agreement with parenting decisions), and Instrumental (help with babysitting or other pa renting responsibilities if needed). These three variables are calculated by summing the 5-point Likert-type responses on the two items constituting each functional subscale. Scores for each subscale can range from 0, if the respondent answers 0 to both s ubscale questions, to 8, if th e respondent answers 4 to both questions. Two composite variables, Emoti onal Support and Total Functional Support, are calculated by combining these subscales. Em otional Support is the sum of Affective and Affirmational functional subscale scores. Tota l Functional Support is the sum of all three subscale scores and is also considered the globa l index of NSSQ caregi ver support. Scores are calculated by summing all ratings for all networ k members rather than averaging scores to capture quantity and quality of support. Norbeck suggests that participants are unlikely to list network members who provide low levels of support, and summing scores more accurately reflects total support (1995). The Caregiver Strain Questionnaire (CGSQ; Brannan and Hefli nger, 2001) is an adult self-report instrument containing 21 items that address the perceived impact on caregivers of children with emotional and/or behavioral problem s. This measure has shown strong reliability and validity in multiple populations. On the CG SQ, caregivers rate how problematic they find various negative experiences or feelings that re sult from their childs pr oblem behaviors. Items such as family disruption, time demands, fina ncial strain, and embarrassment are rated on a 20

PAGE 21

Likert-type scale from 0 ( not at all ) to 5 (very much). Three dimensions of negative caregiver experiences or feelings are measured by the CSQ: Objective (11 items), Internalized Subjective (6 items), and Externalized Subjective (4 items). The Mental Health Inventory-5 (MHI-5; Veit & Ware, 1983) is a brief 5-item measure of adult psychological distress and well-being in ge neral populations. Responde nts rate their degree of psychological distress on a 6point Likert-type scale, and th e sum of these ratings on all 5 items generates a total score. Procedure For this study, parent-report data obtained for youths at high risk for or with pre-existing ADHD diagnoses at Times 1, 2, and 3 were used. Demographic data were obtained from school records and were updated at each assessment point Parents were contacted by phone to arrange a convenient time and place for the assessment, which included both in-home and communitybased locations such as a public library meeti ng room. Assessments were conducted by trained graduate research assistants in clinical or school psychology, and childcare was provided if needed by trained undergraduate students. The a ssessments lasted approximately 90 minutes at Time 1, 60 minutes at Time 2, and 90 minutes at Time 3. Data collection for Time 2 began approximately two years after Time 1 collection but continued up to four years later. Data collection for Time 3 began approximately one y ear after Time 2 collect ion and is currently ongoing. Assessments were conducted by one of three gr aduate research assist ants. Their training included general interviewing skills as well as pr actice in administration of the NSSQ. A script outlining the procedures for test administration was followed to ensure consistency in administration across interviewers. Assessors re ceived approximately 4 hours of instructional training for the NSSQ including role-playing, video-taped practice, and practice administration 21

PAGE 22

with the primary investigator to es tablish inter-rater reliability at or above .90 prior to initial data collection. Inter-rater reliability was reassessed at three-month in tervals, and interviews were supervised randomly to guard against drift from the administration protocol. Data Analysis Data Integrity Data were analyzed using the Statistical Package of the Social Sciences (SPSS). Rating distributions of the NSSQ and CSQ were analyzed to ensure normality of scores. Entry error was checked through visual inspecti on of item ranges for potential outliers using box plots. Item sums were used in analyses as indicated by N SSQ scoring procedures and empirical support of the use of sums over item averages (Norbeck, 1995; Gigliotti, 2002). Reliability Analysis A bivariate correlations matrix and Cronbachs alpha were used to examine internal consistency via inter-item and item-total correlations. Paired samples t tests were first conducted on each item mean score pairing (e.g., Time 1 scores and Time 2 scores) to screen for significant mean differences across time in order to ensure th at any significant test-retest correlations reflect true stability and a non-changing mean over time. Pearson correlation coefficients were calculated for participants who completed the NSSQ at both Time 1 and Time 2 to assess 4-year test-retest reliability ( n = 156). Additionally, 2-year test-ret est correlations between Time 2 and Time 3 were calculated ( n = 49). Correlational magnitude was evaluated using Cohens suggested guidelines of 0.2 as small, 0.5 as medium, and 0.8 as large (1988). Validity Analysis Convergent validity To examine convergent validity, correlational analyses were also conducted to assess the degree to which the NSSQ corresponds with a validated measure of caregiver strain, the 22

PAGE 23

23 Caregiver Strain Questionnair e (CSQ). Convergence between: (a) NSSQ Aid scale and CSQ Objective Strain subscale; (b) NSSQ Emotional Support scale a nd CSQ Subjective Internalized Strain subscale; and (c) NSSQ Total Functi onal scale and CSQ Global Strain scale was examined. The NSSQ was administered concurrently with the CSQ at Time 1, thus Time 1 data were used for these convergent validity analyses ( n = 262). Convergent validity of the Time 2 administra tion of the NSSQ with the Mental Health Inventory-5 (MHI-5) was also analyzed ( n = 200). Pearson correlations were calculated between the MHI-5 Total score and the NSSQ Aid, Affective Support, and Affirmative Support subscales as well as the NSSQ composite scales, Emotional Support and Total Functional Support. Predictive validity To examine the predictive validity of the N SSQ, Pearson correlations were calculated for scores of participants comp leting both the NSSQ at Time 1 and the CSQ, at Time 2 ( n = 138). The same NSSQ and CSQ subscales were examin ed as in the converg ent validity analyses. Additionally, correlations were calculated betwee n scores of mothers completing both the NSSQ at Time 2 and the MHI-5 at Time 3 ( n = 114). Specifically, correlations between the MHI-5 Total score and the NSSQ Aid, Affective Support, and Affirmative Support scales as well as the NSSQ composite scales, Emotional Support a nd Total Functional Suppor t were calculated.

PAGE 24

CHAPTER 3 RESULTS NSSQ data from mothers of children at hi gh risk for or with current ADHD diagnoses were analyzed for this study using the Statistical Package of the Social Sciences (SPSS). Data were screened for normality and judged to ha ve normal distributions. Table B1 present the means, standard deviations, and range of scores for participants who completed the NSSQ at Time 1 and Time 2 ( ntime 1 = 266, ntime 2 = 207, ntime 3 = 49). Reliability Internal consistency was examined using inter-item and item-total correlations as well as Cronbachs alpha. Inter-item and itemtotal correlations for Time 1 ( n = 266) were high with Aid items being the least correlated with other subscales, though remaining highly related. These findings are consistent with previous internal consistency patterns and suggest that Affect and Affirmation items may not represent distinct subscales. Table B2 displays these results. Cronbachs alpha for Time 1 functional item s Affect, Affirmation, and Aid was .60, and as expected, removal of Aid items from calcula tions increased alpha to .84 providing further support that Affect and Affirma tion items are highly correlated and may not be distinctive. Consistent with Time 1 data, Cronbachs alpha for Time 2 ( n = 207) functional items Affect, Affirmation, and Aid was .68. Removing Aid items from calculations again resulted in an increased alpha of .85 for Affect and Affirmation items alone. Paired samples t tests were conducted to assess mean differences between item means at Time 1 and Time 2 as well as Time 2 and Time 3. No significant differences in means were found for any scale or subscale, indicating that the two and fou r-year test-retes t correlations accurately represent item score stability. Table B3 displays the twoand four-year test-retest stability coefficients for NSSQ subscales (i.e ., Affect, Affirmation, and Aid) as well as 24

PAGE 25

composite scales (i.e., Emotional Support and To tal Functional Support) for participants who completed the NSSQ at both Time 1 and Time 2 ( n = 156) and at Time 2 and Time 3 (n = 49). Test-retest reliability between Time 1 and Time 2 NSSQ functional and composite item scores was moderate and ranged between .37 and .45. F unctional and composite item correlations at Time 2 and Time 3 were also moderate and ranged between .30 and .38. Convergent Validity Analyses were conducted to determine the de gree to which the NSSQ is related to a measure of caregiver strain, the Caregiver Strain Questionnaire (CSQ), in the Time 1 sample of 262 mothers of children at high risk for or w ith ADHD diagnoses. Correlations between the NSSQ Aid scale and CSQ Objective Strain su bscale; NSSQ Emotional Support scale and CSQ Subjective Internalized Strain subscale; and N SSQ Total Functional scale and CSQ Global Strain scale are presented in Table B4. The obtained co rrelations were small and not statistically significant, ranging from -.01 to .07. Convergent validity of the Time 2 administra tion of the NSSQ with the Mental Health Inventory-5 (MHI-5; Veit & Ware, 1983) was also analyzed. Pearson correlations were calculated between the MHI-5 Total score and the NSSQ Aid, Affective Support, and Affirmative Support scales as well as the NSSQ composite scales, Emotional Support and Total Functional Support. Correlations were small and not statistically significan t. Table B5 displays these results. Predictive Validity Predictive validity of specific Time 1 NSSQ scales was assessed using specific Time 2 CSQ subscales as the criterion variables (n = 138). The same NSSQ and CSQ subscales examined in convergent validity an alyses were used to examine predictive validity of the NSSQ. Correlations were small and non-significant, su ggesting that social support as measured by 25

PAGE 26

26 NSSQ scales does not predict car egiver strain assessed up to f our years later among mothers of children at risk for or with ADHD. Table B6 displays these results. Predictive validity of the Time 2 administ ration of the NSSQ was also assessed using Time 3 MHI-5 total scores for mothers of childre n at high risk for or with ADHD. Correlations between the MHI-5 Total score and the NSSQ Aid, Affective Support, and Affirmative Support scales as well as the NSSQ composite scales Emotional Support and To tal Functional Support, were small and not statisti cally significant. Table B7 displays these results.

PAGE 27

CHAPTER 4 DISCUSSION Psychometric analyses of the NSSQ in this sample of mothers of children with ADHD symptomatology were not wholly consistent w ith previous psychometric findings, providing only limited support for the use of the NSSQ in this population. Internal consistency of the NSSQ in our sample of mothers was moderate to high with Affective and Affirmation items being more highly related to each other than with Aid items. This finding is consistent with previous studies of the NSSQ in other populatio n samples, lending support for the use of two composite variables, Emotional Support and Ai d, rather than an i ndex of total support. Test-retest reliability of the NSSQ up to a four year interval was low in this sample. However, previous test-retest analyses of th e NSSQ with students and university employees demonstrated high stability of the NSSQ scales when assessed over one week and seven month intervals. Shorter retest intervals are more typica l in psychometric analysis and it is possible that the NSSQ would demonstrate adequate stability for use in ADHD families if examined over shorter spans of time. However, Sarason, Saraso n, and Shearin (1986) found moderate three year test-retest correlations for another measure of social support using a sample of university students. Their findings suggest that perceived support network composition is more stable than satisfaction with the level of support provided. Ther efore, inclusion of al l NSSQ items, such as proximity to support network member s, in test-retest analyses may increase the stability of the instrument across time. For example, the inclus ion of information abou t how physically close a network member is over time is important to und erstanding changes in support such as decreased aid-related support, which would lik ely be negatively related to ge ographic proximity. It is also possible that in our sample there were actual decr eases in the mothers functional support as their 27

PAGE 28

children got older and that the NSSQ was demo nstrating sensitivity to real change over the extended test-retest time intervals. Our analysis of the NSSQ with mothers of children with ADHD symptoms also failed to reveal evidence of convergent validity. Contrary to our hypotheses, there was little convergence of specific NSSQ scales and subscales with CSQ subscales and scales. Previous studies with the NSSQ have shown strong relations between NSSQ Aid items and a measure of life stress. The absence of convergence between th e NSSQ measures of perceived support and caregiver strain as measured by the CSQ was unexpected. The w eak relationship between the NSSQ and an index of overall mental health was also unexpect ed, particularly for mothers of children with ADHD, a group more at risk for depression and significant anxiety than mothers of children without ADHD. Our hypotheses concerning the predictive validit y of the NSSQ, in which we expected social support to be negatively related to careg iver strain and psychologi cal distress over time, were also not confirmed. In light of the absen ce of concurrent relations between the NSSQ and our measure of caregiver strain, it is less surprising to find that the measures do not relate over time. Initial evidence for the predictive validity of the NSSQ has not been replicated and remains important for further examination of this instrument. Previous psychometric study of the NSSQ ha s been based on non-clinical samples of convenience. The social support literature has identified links between social support and parental stress in numerous studies, though the mechanisms of association vary. For example, childhood co-morbidities such as ODD are more str ongly associated with parental stress than singular diagnoses. Financial burden also uniquel y accounts for higher parental stress in mothers of children with ADHD (Johnston & Mash, 2001). Fu ture studies within this population should 28

PAGE 29

account for pre-existing stressors given to the unique effects of child ADHD symptom severity, co-morbid behavior problems, and familial factors unrelated to child behavior on parenting stress (Podolski & Nigg, 2001). Also, research on changes in parental support systems throughout child development, particularly for children with ADHD symptomatology, is no tably lacking but may be influential in longitudina l research involving samples of parents and their children. Future research on the NSSQ may also includ e a revision of the inst rument to include a wider variety of support types. While the measure currently aims to capture the multidimensional nature of support, psychometric analyses of the NSSQ have f ound only two distinct aspects of support are being measured. Other types of suppor t have been noted as important to health outcomes including information support and so cial companionship. Informational support measures ones perception of the receipt of usef ul information such as education, advice, and referral to other sources of support. Social comp anionship refers to participation in leisurely activities with a support network member. The inclusion of ot her types of support may generate a more comprehensive picture of perceived suppor t and reveal a stronger link between social support and current as well as future health ou tcomes (Langford, Bowsher, Maloney, & Lillis, 1997). Limitations of this study in clude the restriction of ava ilable measures for validity analyses due to the secondary analysis of available data. Because anal yses were conducted on existing data, test-retest interval s were longer than conventional in tervals for reliability analysis, which prevented replication of previous findings. Results may have been impacted by method variance and future studies may benefit from employing multiple raters of familial social support. Finally, findings are ba sed on a regional sample and use of geographically broader 29

PAGE 30

samples is needed. Despite the limitations of this study, findings are s upported by a large, welldefined sample that represents the local ethni c and gender distribution among this age group. Features of the NSSQ such as the ability to measure multiple types and sources of support are integral to measures of perc eived social support. The NSSQ employs a multidimensional approach to the measurement of social support, which is necessary for complex constructs. Previous testing of the NS SQ yields support for its use in non-clinical populations. Based on the results of this study, howe ver, further research and refinement of the NSSQ is needed before use in clinical child populations. 30

PAGE 31

31 APPENDIX A NSSQ ITEMS Table A-1. NSSQ items used to rate the quality of social supp ort provided by members of the respondents social support network Items Functional type *1. How much does this person make you feel liked or loved? Affect *2. How much does this person make you feel respected or admired as a parent? Affect *3. How much can you confide in this person? Affirmation *4. How much does this person agree with or support your parenting? Affirmation *5. If you needed babysitting, a ride to the doctor, or some other immediate help, how much could this person usually help? Aid *6. If you were confined to bed fo r several weeks, how much Aid could this person help with your parenting responsibilities? 7. How long have you known this pe rson?* Duration of relationship 8. How frequently do you have contact with this person? Frequency of contact 9. How close do you feel to this person? Closeness of relationship 10. How close does this person live to you? Geographical distance Note. Item responses are rated on a 5-point Likert-type scale for items marked with an asterisk. Item 9 is rated on a 3-point Likert-type scale. Item 10 is rated on a 7-point Likert-type scale

PAGE 32

APPENDIX B ADDITIONAL DATA Table B-1. Time 1 and time 2 means, st andard deviations, range of scores NSSQ Scales Time 1 ( n = 266) Time 2 ( n = 207) Mean SD Range Mean SD Range Affect 27.92 16.13 0-87 27.23 16.56 0-88 Affirmation 27.10 15.69 0-88 26.30 15.58 0-80 Aid 19.31 12.93 0-88 19.27 13.46 0-79 Emotional Support 55.02 31.64 0-173 53.54 31.97 0-160 Total Functional 74.33 42.91 0-255 72.72 43.84 0-238 Table B-2. Pearson correlation coefficients for all pairs of items on time1 NSSQ 1 2 3 4 5 6 7 8 1 Affect1 -2 Affect2 0.98 -3 Affirmational1 0.96 0.96 -4 Affirmational2 0.96 0.96 0.95 -5 Aid1 0.83 0.81 0.80 0.79 -6 Aid2 0.82 0.80 0.78 0.77 0.94 -7 Emotional Support 0.99 0.99 0.98 0.98 0.82 0.80 -8 Total Functional 0.98 0.97 0.97 0.96 0.90 0.89 0.99 -Note. n = 266. All correlations are significant at p < .01 32

PAGE 33

Table B-3. Test-retest stability Pearson correlations for NSSQ scales Four-Year Retest Two-Year Retest Time 1 to Time 2 Time 2 to Time 3 NSSQ Scales ( n = 156) ( n = 49) Affect 0.37** 0.36** Affirmation 0.39** 0.35** Aid 0.45** 0.31** Emotional Support 0.38** 0.38** Total Functional 0.38** 0.30** ** p < .001 Table B-4. Time 1 Pearson corre lations between the NSSQ scales and specific Caregiver Strain Questionnaire subscales ( n = 262) NSSQ Scales CSQ Subscales Objective Strain Subjective-Internalized Strain Global Strain Aid -0.01 --Emotional Support -0.07 -Total Functional --0.05 Table B-5. Time 2 Pearson correlations between the NSSQ scales and the Mental Health Inventory-5 total score ( n = 200) NSSQ Scales MHI-5 total Affect -0.05 Affirmation -0.05 Aid -0.08 Emotional Support -0.05 Total Functional -0.06 33

PAGE 34

34 Table B-6. NSSQ Scales Predictive Validity of Specific CSQ Subscales ( n = 138) NSSQ Scales CSQ Subscales Objective Strain Subjective-Internalized Strain Global Strain Aid 0.00 --Emotional Support -0.06 -Total Functional --0.04 Table B-7. NSSQ Scales Predic tive Validity of Mental Health Inventory-5 Total Scores ( n = 114) NSSQ Scales MHI-5 Total Time 2 Time 3 Affect 0.04 Affirmation 0.01 Aid -0.05 Emotional Support 0.03 Total Functional 0.01

PAGE 35

LIST OF REFERENCES American Psychological Association. (1994). Diagnostic and statistica l manual of mental disorders: DSM-IV (4th ed.). Washington, DC: Author. Brannan, A. and C. Heflinger, (2001). Disti nguishing caregiver strain from psychological distress; modeling the rela tionships among child, famil y, and caregiver variables Journal of Child and Family Studies, 10, 4, 405-418. Brandt, P.A., and Weinert, C. (1981). The PRQa social support measure. Nursing Research, 30, Sept-Oct., 277-280. Bussing, R., Zima, B.T., Gary, F.A., Mason, D.M ., Leon, C.E., Karabi S., Garvan, C.W. (2003). Social networks, caregiver st rain, and utilizatio n of mental health services among elementary school students at high risk for ADHD. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 7, 842-850. Carney, J., Boyle, M., Offord, D.R., & Raccine, Y. (2003) Stress, social support, and depression in single and married mothers. Society of Psychiatry and Psyc hiatric Epidemiology, 38, 442-449. Cohen, J. (1988). Statistical power analysis for the behavioral sciences Hillsdale, New York: Lawrence Erlbaum Associates. Cohen, S., and Hoberman, H.M. (1983). Positive ev ents and social supports as buffers of life change stress. Journal of Applied Social Psychology, 13, 99-125. Colletta, N. (1979). Support systems af ter divorce: incidence and impact. Journal of Marriage and the Family, 41(Nov, 4), 837-846. Crowne, D.P. and Marlowe, D. (1960). A new s cale of social desirabi lity independent of Psychopathology. Journal of Counseling Psychology, 24, 349-354. Dadds, M., & McHugh, T. (1992). Social support a nd treatment outcome in behavioral family therapy for child conduct problems. Journal of Consulting and Clinical Psychology, 60(2), 252-259. Devall, E., Stoneman, Z., & Brody, G. (1986) The impact of divorce and maternal employment on pre-adolescent children. Family Relations, 35, 153-159. Colletta, N.D. (1979). Support systems af ter divorce: Incidence and impact. Journal of marriage and the family, 41 4, 837. Coventry, W.L., Gillespie, N.A., Heath, A.C., & Martine, N.G. (2004). Perceived social support in a large community sample. Social Psychiatry & Psychia tric Epidemiology, 39, 625636. 35

PAGE 36

Dormire, S.L., Strauss, S.S., & Clarke, B.A. ( 1989). Social support and adaptation to the parent role in the first-time adolescent mothers. Journal of Obstetric, Cynecologic, and Neonatal Nursing, 18(4), 327-336. Eyberg, S., Boggs, S., & Rodriguez (1992). Relations hips between maternal parenting stress and child disruptive behavior. Child and Family Therapy, 14 (4) Fischer, M. (1990). Parenting Stress and the Child With Attention Deficit Hyperactivity Disorder. Journal of Clinical Child Psychology, 19(4), 337-346. Gigliotti, E. (2002). A confirma tion of the factor structure of the Norbeck Social Support Questionnaire. Nursing Research, 51, 276-284. Gigliotti, E. (2006). A confirmatory factor anal ysis of situation-specifi c Norbeck Social Support questionnaire items. Nursing Research, 55, 3, 217-221. Hanson, S. (1986) Healthy single parent families. Family Relations, 35, 125-132. Hilton, J.M., Desrochers, S., & Devall, E.L. (2001 ). Comparison of role demands, relationships, and child functioning in single-mother, single-father, and intact families. Journal of Divorce & Remarriage, 35, 29-55. Hollingshead, A.B. (1975), Four Factor Index of Social Status. New Haven, CT: Yale University, Department of Sociology. Johnston, C., and Mash, E. (2001). Families of ch ildren with attention-hyperactivity disorder: Review and recommendations for future research. Clinical Child and Family Psychology Review, 4 (3), 183-207. Kahn, R.L. (1979). Aging and social support. In M.W. Riley (Ed.), Aging from birth to death: Interdisciplinary perspectives, 77-91, Boulder, CO: Westview Press. Lakey, B., Adams, K., Neely, L., Rhodes, G., Lu tz, C., & Sielky, K. (2002). Perceived support and low emotional distress: The role of en acted support, dyad similarity, and provider personality. Personality and Social Psychology Bulletin, 28, 1546-1555. Lange, G., Sheerin, D., Carr, A., Dooley, B., Ba rton, V., Marshall, D., Mulligan, A., Lawlor, M., Belton, M., & Doyle, M. (2005). Family factors associated with attention deficit hyperactivity disorder and emo tional disorders in children. Journal of Family Therapy, 27, 76-96. Langford, C. P., Bowsher, J., Maloney, J., Lillis, P. (1997). Social support: a conceptual analysis. Journal of Advanced Nursing, 25, 95-100. McCurdy, K. (2005). The influence of suppor t and stress on maternal attitudes. Child Abuse & Neglect, 29 (3) 251-268. 36

PAGE 37

McNair, D.M., Lorr, M., & Doppleman L.F. (1971). POMS Manual for Profile of Mood States. San Diego, CA. Educational and In dustrial Testing Service, 1971. Miller P., McMahon, M.A., & Garrett, M.J. (1989) A content analysis of life adjustments post infarction. Western Journal of Nursing Research, 11(5) 559-567. Norbeck, J. (1995). Scoring Instructions for the Norbeck Social Support Questionnaire (NSSQ). University of California, San Francisco. Norbeck, J.A., Lindsey, A. M., & Carrieri, V.L. (1981). The development of an instrument to measure social support. Nursing Research, 30, 5, 264-269. Norbeck, J.S., Lindsey, A.M, & Carrieri, V.L. (1982). Further Development of the Norbeck Social Support Questionnaire: No rmative Data and Validity Testing. Nursing Research, 32, 1, 4-9. Orth-Gomer, K. and Unden, A. (1987). The me asurement of social support in population surveys. Social Science a nd Medicine, 24(1) 83-94. Podoloski, C.L., & Nigg, J.T. (2001). Parent stress and coping in relation to child ADHD severity and associated child disruptive behavior problems. Journal of Clinical Child Psychology, 30, 503-515. Rodgers, A. (1998). Multiple sources of stress and parenting behavior. Children and Youth Services Review, 20 (6), 525-546. Rice, C., & Longabaugh, R. (1996). Measuring genera l social support in alcoholic patients: short forms for perceived social support. Psychology of Addictive Behaviors, 2, 104-114. Sandler, I.N., & Barrera, M., Jr. (1984). Toward a multimethod approach to assessing the effects of social support. American Journal of Community Psychology, 12, 37-52. Sarason, I.G., Johnson, J.H., & Siegel, J.M. (1978). Assessing the impact of life changes: Development of the Life Experiences Survey. Journal of Consulting Clinical Psychology, 46, 932-946. Sarason, I.G., Sarason, B.R., & Shearin, E.N. (1986). Social support as an individual difference variable: Its stability, origin s, and relational aspects. Journal of Personality and Social Psychology, 50(4) 845-855. Sarason, B.R., Shearin, E.N., Pierce, G.R., & Sarason, I.G. (1987). Interrelations of social support measures: Theoretical and practical implications. Journal of Personality and Social Psychology, 52 (4), 813-832. Schaefer, C., Coyne, J., & Lazarus, R. (1981). The health-related functions of social support. Journal of Behavioral Medicine, 4 (4) 381-406. 37

PAGE 38

38 Suarez, L.M. and Baker, B. L. (1997). Child extern alizing behavior and parents stress: the role of social support. Family Relations, 47, 373-381. Tietjen, A.M. (May, 1985). The social networ ks and social support of married and single mothers. Journal of Marriage and the Family, 489-496. Treacy, L., Tripp, G., & Baird, A. (2005). Pa rent stress management for attentiondeficit/hyperactivity disorder. Behavior Therapy, 36, 3, 223-233. Wahler, R. (1980). The insular mother: her problems in parent-child treatment. Journal of Applied Behavior Analysis, 13, 207-219. Wahler, R. G., & Dumas, J. E. (1989). Atten tional problems in dysfunctional mother-child interactions: An interbehavioral model. Psychological Bulletin, 105, 116-130. Winemiller, D. R., Mitchell, M. E., & Sutliff, J. (1993) Measurement strategies in social support: A descriptive review of the literature. Journal of Clinical Psychology, 49, 5 638-648. Wan, C., Jaccard, J., & Ramey, S. (1996). The relationship between so cial support and life satisfaction as a function of family structure. Journal of Marriage and the Family, 58, May, 502-513.

PAGE 39

BIOGRAPHICAL SKETCH Monica Leah Stevens was born in Laurel, Mi ssissippi in 1984. She was raised in Mississippi along with her older and younger brother and graduated from the Mississippi School for Mathematics and Science in May of 2002. She then enrolled in the Un iversity of Southern Mississippi Honors Coll ege and graduated summa cum laude with a Bachelor of Science degree in psychology in 2006. In 2006, Monica enrolled in a dual Master of Science and Doctor of Philosophy program in the University of Floridas Department of Clinical and Health Psychology. She currently has maintains an assistantship in the Child Study Lab under the mentorship of Sheila Eyberg and Stephen Boggs Upon completion of her Master of Science, Monica continued her Doctor of Philosophy work in the Department of Clinical and Health Psychology. 39