|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
MEASURING MORPHOLOGICAL AWARENESS IN ADULT READERS: IMPLICATIONS
FOR VOCABULARY DEVELOPMENT
LISA KAY MAAG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2007 Lisa Kay Maag
In memory of my parents
I thank my family and friends for their support during this endeavor. I also thank my
committee members for their patience and their advice, especially my Committee Chair, M.
Jeffrey Farrar. In addition, I want to thank all of the staff of Project Care, particularly Fonda
Eyler, Marylou Behnke, Kathleen Wobie, and Ann Welch, for making my time at the University
of Florida rewarding and enjoyable.
TABLE OF CONTENTS
A CK N O W LED G M EN T S ................................................................. ........... ............. .....
L IS T O F T A B L E S .............................................................................. ............... 7
LIST OF FIGURES .................................. .. ..... ..... ................. .8
A B S T R A C T ................................ .................. .......................... ................ .. 9
1 O V E R V IE W .................................................... ....................... ................ 10
2 INTRODUCTION ............... ............................ .............................. 14
Estim ates of V vocabulary Size ........................................................... .. ............... 14
Theories of V ocabulary D evelopm ent........................................................................ ...... 17
The B asics of M orphology .................................................................................. 23
Exposure to Words: The Language Environment ................................. ...............25
The Distribution of W ords: W ord Frequency ............................................................. 26
Where the Big Words Are: Why Reading Experience is Necessary..............................28
T im e S p en t R ead in g .................................................................................... ................. .. 3 0
M orphology and Spelling ............................................................................. .................... 32
M measuring M orphology Skill........................................................................... .............. 33
Previous Studies of M orphology Skill ....................................................... ...............33
Making the Case for Morphological Awareness..........................................................38
The Rationale for the Morphological Awareness Measure...........................................42
H ypotheses.......... ...........................................................44
3 M E T H O D ..........................................................................4 6
P articip an ts .........................................................................4 6
In stru m e n ts ........................................... ... ....................... ..............................4 6
A Standardized Reading Test: The Nelson-Denny Reading Test ................................46
The Morphology Test Designed for the Current Study ...........................................47
Scoring ..................................................................................... 48
P ro c e d u re .............. .... ...............................................................4 9
4 R E SU L T S .............. ... ................................................................53
Amount of Reading ...........................................................................53
Participants' R leading B background ............................................. ................. 54
Individual T ests .............................................................................55
N elson-D enny R leading Test................................................................. 55
M orphology Test: Subtests 1, 2, and 3 .........................................................................55
Self-assessm ent of Reading and V ocabulary................................... .................................... 58
Comparison of Better and Less-skilled Readers................................. ....................... 58
5 D IS C U S S IO N ........ .. ........................................................................................6 7
F findings of the C current Study ......................................................................... .................. 67
Reasons to Pursue Morphological Awareness Research..................................................72
APPENDIX: MORPHOLOGY TEST .......................................................... ...............76
L IST O F R E F E R E N C E S .............................................................................. ...........................84
B IO G R A PH IC A L SK E T C H .............................................................................. .....................93
LIST OF TABLES
3-1 Languages participants use fluently. ............................................. ............................. 51
4-1 Amount of reading per week during the school semester .................................................59
4-2 Nelson-Denny Reading Test scores for the standardization sample and University of
F lorid a stu d ents............................................................................................... 6 0
4-3 Performance on morphology test subtests and metalinguistic measures (n = 106) ..............61
4-4 Intercorrelations am ong m measures. ............................................... ............................... 62
4-5 Correlations of Nelson-Denny and MA (Total and hard words) scores ..............................63
4-6 Linear regression predicting Nelson-Denny Total Reading (raw scores) from three
predictor variables................................................................. ..... ..........64
4-7 Performance of good and less-able readers................................ ........................ ......... 65
4-8 Numbers and percentages of participants in each quartile of Nelson-Denny scores ..............66
LIST OF FIGURES
3-1 Scoring of the m orphology test. ............................................................................. ............52
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
MEASURING MORPHOLOGICAL AWARENESS IN ADULT READERS: IMPLICATIONS
FOR VOCABULARY DEVELOPMENT
Lisa Kay Maag
Chair: M. Jeffrey Farrar
Morphological awareness (MA) is considered a prime force in children's vocabulary
growth, but this skill has scarcely been measured in adults. This study tested a method for
assessing MA using complex (multimorphemic) words and a two-stage process designed to elicit
explicit morphological knowledge. Participants, 106 college students, indicated which of the
complex words they knew and were tested on morphological analysis and on definitions of the
words. For example, correct responses on the morphology test would indicate that readmission is
related to admit (and not to mission) and that presumptuous is based onpresume (and not
sumptuous). A standardized test, the Nelson-Denny Reading Test (ND), was also administered.
Two metalinguistic scores were derived: MA was calculated for known words for which the
morphological base word was identified, and Accuracy was based on known words for which the
correct definition was selected. Participants with better ND Vocabulary and Comprehension
scores also had higher MA scores and greater metacognitive awareness of their own lexicons. An
inadequate vocabulary is one cause of reading problems, and a method that extends MA research
to an older age group allows further research into vocabulary development with implications for
In the process of growing up in a literate society, most young people gain a large
vocabulary, and word learning can continue throughout the lifespan. It is hard to appreciate the
scale of this achievement because no one knows precisely how many words are learned. There is
an enormous research literature describing the early years of language development, and many
mechanisms of word learning have been identified in young children. There is also a large
research base about learning to read and the related gains in language ability during early years
of schooling. Most of the research has focused on how communication ability develops in
childhood, and less is known about vocabulary development in adolescents and adults. There is
still no complete explanation for the vast word-learning accomplishment of the first 15 or 20
years of normal development.
In English, most words in a child's vocal repertoire are of Anglo-Saxon origins, such as
house, hand, good, walk, cold, and who. The Anglo-Saxon words are usually only one or two
syllables long, and these simple words must be learned individually. A child may learn
something about a word from hearing it once, but word learning is often extended, and
knowledge of words is built up incrementally. Hearing or reading a word in an informative
context can generate some learning, but many words may require multiple exposures to provide
the young learner with sufficient understanding.
In school, as the child encounters more sophisticated language, big words become more
common. Many big words are of Latin and Greek origins, such as eradicate, complementary,
paradigm, and consequential. Many of these big words are formed through derivations: act
(verb) can become active (adjective) or action (noun), and these lead to transaction, activation,
proactive, reactionary, and more. The prefix trans- is also used in transport, which is related to
import, export, deportation, and deportment. The suffixes -ation and -ment, while difficult to
define, indicate a state or condition of something, and these two suffixes signal nouns. This
sharing of morphemes among derived words creates a vast network of word relationships, and
this quality means that many of these words do not have to be learned separately. New words
made of familiar parts might be understood spontaneously, or a new word may be at least
partially understood if some of the morphemes are recognizable. For example, a student who
knows the words biology and geology, might not know the root of seismology, but would
probably realize that the word is a noun and refers to some area of study. In this way, learning
complex words provides leverage that assists in additional vocabulary gains.
Everyday speech tends to use the frequent words that are known by almost all native
speakers, while the longer, derived, less common words occur more often in formal, written text.
As students move into fourth grade and beyond, the vocabulary of textbooks increases in
complexity and difficulty. As these vocabulary demands increase with increasing grade level, the
more difficult words may also include helpful clues-but only for the skilled language users.
Poor readers who have not gained sufficient experience with derivational morphemes may
struggle to master the new words.
A thorough knowledge of vocabulary depends on concrete knowledge of specific words,
plus the understanding of the connections that exist among derivationally-related words. Most
conventional vocabulary tests measure the grasp of word meanings, and tests that assess the
understanding of morphology are seldom used, except in some research studies.
Two highly influential studies have examined the conditions of English vocabulary-
learning, and these papers prompted the current study. In the first study, Nagy and Anderson
(1984) analyzed the words found in typical school materials through grade 9, and determined
that the teaching of vocabulary would fall short because there are too many words to be taught.
They concluded that children must be gaining much of their vocabulary through reading, and by
making inferences about words that are semantically transparent.
Anglin (1993) interviewed school children to learn about the depth and breadth of their
word knowledge. The children were encouraged to try to explain words that they were unable to
define. For example, one child who didn't know the word treelet, decided by analogy that it must
mean a small tree because a piglet was a little pig. Anglin called this process "morphological
problem solving" (p.5) and showed that understanding of multimorphemic words was low in first
graders, but increased dramatically by fifth grade. He suggested that the ability to analyze the
morphemes in complex words accounted for a large part of vocabulary growth.
In addition, a number of studies have examined children's ability to use morphemes, and
the data show greater understanding of morphology at higher grades, with correlations between
morphology skill and reading ability. An increase in morphology skill should serve both as a
cause of and an effect of better reading. If this skill continues to improve with advanced reading,
older students should have higher levels of morphological skill that can be useful in
comprehending advanced text. Unfortunately, this area of research has not yet been extended
past the middle school level, and the testing methods currently being used with children are too
easy for older readers.
The goal of the present study was to investigate the morphological ability of young,
(college-age) adults. If the analysis of morphology propels vocabulary growth, young adults
should have morphological ability that matches their reading ability-skilled readers should have
sophisticated morphology skills.
As another consideration, this study was designed to test the explicit knowledge of
morphology. The importance of metacognitive skills to reading has been proposed, but has not
had much testing other than in the area of phonological awareness. There have been studies
showing the value of self-monitoring of comprehension during reading, but other metalinguistic
skills have received little attention in research on reading. Phonological awareness refers to the
ability to explicitly reflect on and manipulate phonemes, and this skill is a key to successful
reading acquisition. Once the basic reading processes have been mastered, the importance of
phonological awareness may decline. Other skills may be essential for advanced reading, and
morphological awareness (MA) is one likely candidate.
To tap explicit awareness of morphology at adult levels of skill, a two-step method was
used. First, the study participants were asked to look over a list of complex words and to check
the words they knew. The second step asked them to identify the morphologically simpler words
from which the complex words might have been formed, and a multiple-choice format was used.
As examples, a person with good MA would know that that readmission is related to admit (and
not to mission), and that presumptuous is based on presume (and not sumptuous). A third section
tested the participants' knowledge of the meanings of the complex words. A standardized
reading test was also administered, and correlations between the new test and the standardized
test were used to assess the new method. The positive correlations between the reading test
norms and the morphological test suggest that this area deserves more study in the quest to
explain vocabulary development.
Estimates of Vocabulary Size
Infants enter the world already recognizing their mother's voice, and with a preference for
that particular language, but it takes around one year for a baby to begin to produce words. From
the single-word stage, it may be another few months to a year before two-word phrases appear.
By the age of two, a child can know hundreds of words, and many children will be speaking in
sentences. The spoken words comprise the toddler's productive vocabulary, and the receptive
vocabulary (words that are understood) is even larger. After breaking into language gradually,
most children make huge gains in vocabulary throughout the first few years of life.
Measuring an individual's vocabulary might seem simple, but it is very difficult to
accurately assess vocabulary knowledge. There have been many estimates of vocabulary size for
different ages, and the totals vary tremendously. At 12 months of age, when most children are
beginning to produce single words, some children have a vocabulary of 52 words and others
have zero (Bates et al., 1994). A typical first-grader may know 2,700 words (Dolch, 1936) or
26,000 (Shibles, 1959). With older students, the estimates diverge even more. An average
college student may have a vocabulary of 17,000 words (Goulden, Nation, & Read, 1990;
D'Anna, Zechmeister, & Hall, 1991), or 200,000 words (Hartman, 1946 cited in Anderson &
Freebody, 1981).These disparities are partially due to actual large-scale differences in
vocabulary size, but they also reflect different approaches to the measurement of word
There are many factors that complicate the measurement of vocabulary. Given a checklist,
a parent can describe an infant's entire vocabulary in only a few minutes. By the age of 3, a child
may say as many as 20,000 words per day (Wagner, 1985), and recording the child's entire
vocabulary may no longer be feasible. Due to the breadth of vocabulary after age 2 or 3, most
measures of vocabulary size rely on sampling procedures, from which the total vocabulary can
The most fundamental problems include defining what is meant by 'a word,' and
determining what constitutes word knowledge (Lorge & Chall, 1963). Even dictionary editors do
not agree on how to define "a word," and different publishers use somewhat different criteria for
determining the main word entries. Most researchers and dictionary editors count a basic word
and its regular inflections as one word, such as look, looked and looking (but for a dissenting
view, see Miller and Wakefield, 1993). Irregular forms of words, such as go and went, are often
counted as separate entries, but some dictionaries list irregular verbs under the basic entries.
Derived words are usually counted as separate entries in dictionaries, but vocabulary tests vary in
the treatment of derivations. Some vocabulary estimates would count happiness, unhappiness,
and happily as one word, while others would count each variation as a separate word.
Homographs and compound words add additional problems. Bear can mean a large mammal or
to carry, and this could be counted as one word or two words. Such decisions have a tremendous
impact on any attempt to count words.
Another problem in measuring a person's vocabulary is that there is no simple definition of
what it means to know a word. Supplying or recognizing a synonym or definition is often used as
the criterion for knowing a word, but meaning is only one aspect of word knowledge (Nagy &
Scott, 1999). Testing knowledge of word meanings is complicated, especially for words with
One method of conceptualizing word knowledge used a scale with 4 levels (Dale, 1965). A
person might (1) never have seen the word before; (2) know that it is a word, but not know the
meaning; (3) know the general idea of a word; or (4) know a word well. Durso and Shore (1990)
described a continuum of word knowledge, anchored by 2 endpoints: unknown words, or those
that had never been encountered, and known words, which are part of a person's productive
vocabulary. In between stretches a landscape of words ranging from slightly known to fairly-
well-known words. Some of the most detailed characterizations of word knowledge have been
devised in the area of second language learning. One classification system specifies nine aspects
of knowledge of a word, including categories of receptive and productive knowledge about the
word's form, its meaning, and its use (Nation, 2001). These various models show that there is no
single, precise definition of knowing a word. Even at the lowest levels of knowledge, word
understanding can be difficult to pin down. A person with only a very vague understanding of a
word's semantic content might still be able to select a correct definition on a test. Conversely, a
person might possess a fairly thorough knowledge of a word, but not recognize the particular
definition used on a test. Few standardized vocabulary instruments investigate depth of word
knowledge, and most simply test for some degree of knowledge of the most common meaning.
Each method of measuring vocabulary involves trade-offs between efficiency and usefulness of
the results, and the chosen testing method will depend on the purpose of the test and the available
These fundamental issues in vocabulary measurement suggest that reports of absolute
vocabulary size must be viewed with caution. Different measures of vocabulary cannot be
directly compared unless they used similar definitions of a word to measure vocabulary breadth,
and used similar criteria for knowing a word to determine depth of word knowledge. It is simpler
to assess relative vocabulary, comparing performance on specific vocabulary tests, or to measure
particular types of vocabulary knowledge.
Measuring knowledge of root or basic words is one way to avoid the word definition
problem. Root words or basic words are not inflected or derived, and must be learned
individually. For example, a child would have to learn the meaning of think, but would be
assumed to know the inflected forms: thinks, thinking, and even the irregular past tense, thought.
By one account, children older than second-graders can gain about 1,000 basic words per year
(Biemiller & Boote, 2006). Using basic words to estimate vocabulary size is a sensible
accommodation to the problem of defining separate words, but such a count does not fully
indicate the depth or quality of a person's word knowledge. To make better use of the basic word
measure, an additional measure would be valuable: the extent to which the word learner can use
morphological clues to interpret complex words formed from basic words. Such a measure
would indicate recognition of word families, as in knowing think and understanding its
derivations: thinkable, thoughtful, thoughtless, unthinkingly, forethought, afterthought, rethink,
methinks, overthink, think tank, and groupthink. This is the ability that would be rated with a test
of morphological awareness (MA).
Due to the methodological problems, we do not know precisely how large an adult's
vocabulary is, but we know that the numbers are large. Accounting for the rapid increase in word
knowledge that begins in childhood and continues throughout adulthood is even more complex,
and likely involves multiple learning strategies.
Theories of Vocabulary Development
Several theories of language development offer explanations for the rapid expansion of
children's language competence. One explanation for the linguistic ability of infants is that they
possess innate principles, constraints, or biases that guide word learning. For example, children
often assume that a new word applies to something for which they did not already have a name.
This is called the mutual exclusivity principle (Markman, 1990), or the novel name-nameless
category principle (Golinkoff, Shuff-Bailey, & Olguin, 1995). A variation is the principle of
contrast that states that no two words mean exactly the same thing (Clark, 1993). The whole
object bias is another principle that would simplify word-to-referent mapping, and a taxonomic
bias would guide the extension of the word to similar objects (Markman, 1990). Using these
guidelines, a child would assume that the word cat refers to the entire animal, and not just to its
fur, its tail, or some other feature, and that cat also applies to other creatures that look like a cat.
Some constraints theorists have maintained that these are innate properties, but others have
proposed developmental pathways for constraints. For example, one set of multiple constraints
has been proposed, beginning with three basic principles that get word learning off the ground,
and evolve into three more sophisticated principles (Golinkoff et al., 1995).
Regardless of the source of these word-learning processes, they are generally thought to be
specific to language acquisition. Other language researchers insist that such principles or
constraints are unwarranted. One criticism is that constraints tend to be invoked more for noun
learning than for other classes of words. This implies that other guidelines are needed for other
word classes and the child would need to know when to apply which guidelines (Nelson, 1988).
Some argue that general learning processes are sufficient to explain vocabulary gains and
that dedicated word learning mechanisms are unnecessary. The shape bias is another principle
that enables children to apply a newly learned noun to other objects with similar forms, but
Smith (1999) describes this as an emergent, not innate, mechanism. Markson and Bloom (1997)
showed that 3- and 4-year-olds learned novel facts as readily as novel words, concluding that a
dedicated word learning system is unneeded. Similarly, Samuelson and Smith (1998) explained
examples of word learning through memory and attention, asserting that specialized skills for
words are unjustified.
Another argument against constraints revolves around the familiar, recurrent routines in
which adults and young children engage. Ritualized behavior sequences help children recognize
actions and objects and link them as referents of the adults' words. Studies of joint attention have
shown the value of adults noticing and describing the focus of the child's attention. For example,
when children were videotaped at 15 and 21 months in play sessions with their mothers, more
conversation occurred during episodes of joint attention than outside of these episodes
(Tomasello & Farrar, 1986). A positive correlation was found between maternal references to the
object of joint attention and the child's language at 21 months.
Different parental styles of speech have also been differentially associated with patterns of
word acquisition. Mothers' speech to their 13-month-olds was coded both for pragmatic intent
and whether or not each statement followed the infant's focus or introduced a different object of
focus (Akhtar, Dunham, & Dunham, 1991). Only the prescriptive commands that followed the
infants' attention had a large positive correlation with the infant vocabulary measure at 20
months. Mothers' statements that redirected their children's attention were not correlated with
later language skills.
The child is not merely a passive recipient of adults' word instruction. Although labeling is
commonplace (at least among middle-class, English-speaking parents), most words are probably
learned without explicit teaching (Tomasello & Barton, 1994). According to the social-pragmatic
view, the child uses an understanding of the communicative intentions of others to identify a
word's meaning (Tomasello, 1995). By 18 months, children have some understanding of others
as intentional agents, and the joint engagement of adult and child is no longer as critical to
semantic gain. The child's growing understanding of others allows insight into word referents
that may not be made explicit. In one study, children 19 to 20 months of age heard an adult use a
novel word (It's a modi!) while looking at a toy (Baldwin, 1993, p. 836). Half of the children (in
the coincide condition) were given the toy immediately after the adult's exclamation. In the
conflict condition, children were given a different toy during a 10s delay, before they were
handed the toy that the adult had labeled the modi. In both conditions, the children were
relatively successful at identifying the object called a modi, demonstrating that they noticed and
correctly interpreted the adult's intentions and were not using associative learning to link an
object and label.
Overhearing words and inferring their meanings may contribute a substantial portion of
word learning (Akhtar, Jipson, & Callahan, 2001). By the age of two-and-a-half, children learned
action verbs or names for objects equally well whether they overheard the word or were spoken
to directly. Slightly younger children (with a mean age of 25 months) could learn nouns through
overhearing them, but were less successful with action verbs.
There is ample evidence that social intentions provide very powerful clues to word
meaning. Even when distracted by a more interesting, remote-controlled toy, 2-year-olds were
able to learn the name for a less salient object that was labeled by an adult. In other words, social
cues had more influence than an attention-grabbing toy (Moore, Angelopoulos, & Bennett,
1999). Still, an overreliance on social-pragmatic information as the one-and-only path to word
meaning may be unwise (Baldwin & Moses, 2001; Hoff & Naigles, 2002). Many autistic
children who seem to lack social referencing skills manage to acquire an adequate vocabulary
(Baldwin & Moses, 2001). Another problem is that joint attention may not occur frequently
enough to be a rich source of input for infants (Hoff & Naigles, 2002). In the 'data-providing
account' of Hoff and Naigles, the primary source of semantic information shifts as a child learns
more about language (2002). For young children, the mother's sensitivity to the child's attentional
focus and provision of appropriate cues may supply the best entry into word learning. By 18
months, children usually have sufficient understanding of others' intentions to figure out many
word meanings on their own. As children's linguistic knowledge increases, the information from
conversations may become the predominant source of new semantic understanding. That is, by
24 months, children can use syntactic and lexical clues to learn about unfamiliar words, and the
greater the conversational richness and quantity, the more information the child has to work with.
In English, word order also provides many clues about meaning (Gleitman, 1994). For
example, when preschoolers heard the sentence "this dog is named Daxy" they interpreted the
novel word as a proper name (Hall & Graham, 1999, p.82). When the word appeared in the
sentence "this dog is very daxy" it was assumed to be an adjective, referring to a quality of the
dog. The 4- and 5- year-olds were willing to let two adjectives refer to the same dog, but not two
proper names, showing that they understand something about the functions of these two word
Verbs may be more difficult to learn than object labels, and verbs are seldom explicitly
taught as are nouns (Gleitman, 1994). Verbs play an important role in the shift from one word
speech to multiword utterances (Olguin & Tomasello, 1993). By the age of 2, children can use
new nouns flexibly, suggesting that they have an apparent category for count nouns. Before age
3, verbs are acquired individually and children use them only in restricted forms. Eventually
children acquire sufficient experience to abstract patterns of verb use and develop a syntactic
category for verbs. By the age of 3 or 4, most children readily add new verbs to their
vocabularies and produce them in appropriately inflected forms, implying that they have an
abstract category of verb. Overgeneralizations of irregular verbs, for example, rarely occur
before this age (Tomasello, 2001).
Other types of language knowledge contribute to learning word meanings. Linguistic
information within an utterance can help a child determine the meaning of an unfamiliar word.
Two-year-olds can use their understanding of familiar verbs (such as eat, ride, or spill) and their
general knowledge to infer the meanings of unfamiliar nouns (Goodman, McDonough, &
Brown, 1998). Children listened to sentences such as "Mommy feeds the ferret," and were asked
to pick out the target noun ferret from a page with drawings of 3 inanimate objects and 1 small
mammal. The 24- and 30-month old participants correctly identified a novel noun an average of
4.3 times out of 10. Thus, by the age of 2 or 3, children can draw on an array of cues to figure
out the meanings of new words.
Young children have many strategies for figuring out new words and use multiple methods
for doing so (Hirsh-Pasek, Golinkoff, & Hollich, 2000; Akhtar, 2002, Bloom, 2000). It is
probably safe to assume that older children can use the same methods and apply them more
skillfully. We know that preschoolers have social referencing skills and can use others' intentions
to determine word meanings. By the age of 4 or 5, children have developed a more sophisticated
theory of mind, so that their understanding of others' sources of knowledge can help them make
inferences (Baldwin & Moses, 2001). Preschoolers have much more general knowledge than
toddlers to help them narrow down the possible referents of words. They are more experienced
with complex syntax and can take advantage of contrasts and cues within sentences. Older
children possess a considerable vocabulary and can use known words to figure out novel words.
These provide a broad repertoire of skills that can be applied to learning the meanings of novel
As children reach school age, the research has focused more on reading skills than on oral
language skills. Although a substantial amount of learning from oral conversation undoubtedly
continues during the school years, there is a dearth of evidence on this topic (Beck & McKeown,
1990). The connections of oral language skills to reading skills have only recently begun to be
investigated in detail (Ouellette, 2006). Most theories of language development have focused on
explaining the mechanisms by which young children gain a foothold in language, and theories of
reading address the development of reading at later ages. There are, as yet, few solid links
between these two theoretical areas. A closer examination of vocabulary and morphology skill
can contribute to the understanding of language development from childhood into adulthood.
The Basics of Morphology
Morphology plays an integral part in the formation and comprehension of English words,
and yet morphemes seem underappreciated. Morphemes, which include roots, stems, prefixes,
and suffixes, are the smallest parts of words that carry meaning. They are usually larger than
phonemes and smaller than words. Although we may think of words as being composed of
letters, it is more accurate to note that words are spelled with letters for the purpose of putting
sound into written form. Words are actually composed of morphemes, which link sound and
meaning. Being able to utilize this intermediate level of language is central to building a
vigorous vocabulary and comprehending English text.
Morphology refers to the use of or the study of morphemes, the components of words that
convey meaning. The precise role of morphology varies with each language, depending on the
word formation processes used in each language. It is not even certain that there is a universal
role of morphology that applies to all languages (Libben & Jarema, 2004), and in this paper, the
focus will be limited to English.
Monomorphemic words (such as window, sleep, and elephant) are also called root words
or basic words and must be learned individually. Multimorphemic words, such as autobiography,
distrustful, and i 1 i/thinu,,, are composed of concatenated morphemes. The term complex word
is often used in vocabulary studies as shorthand for multimorphemic word. In the words talking,
talkative, backtalk, and double-talk, the root word talk is a shared morpheme that links the words
semantically. Root words and stems are modified by inflectional and derivational morphemes
that convey syntactic and semantic details. Inflectional changes reflect grammatical functions of
words, with changes of tense, person, and number. Inflections do not change a word's class, or
part of speech, as in the examples book/books, walk/walks/walked/walking. Changes in
derivational morphology allow a root word or stem to occupy various word classes, as in know
(verb), knowingly (adverb), knowledge (noun), and knowledgeable (adjective). Derivational
changes usually create longer words and are generally more complex than inflectional processes.
Morphemes supply the raw materials for creating new words, and the agility of morphemes
provides part of the generative power of the English language. Many new words are readily
understandable because they are composed of familiar morphemes, such as these additions to a
popular dictionary in 2006: biodiesel, ringtone, spyware, and agritourism (Merriam-Webster,
2006). Noticing the root word inside a complex word, and recognizing prefixes and suffixes are
strategies for figuring out novel words, and skilled readers seem to do this effortlessly and
automatically. With a healthy working knowledge of English morphology, a person can learn
new words 'on the fly.'
Recognizing and understanding morphemes simplifies the child's job of word learning.
Nagy and Anderson (1984) estimated that for each word learned, a child could understand three
others. If every single word had to be learned independently, the vocabulary-learning task would
be much more cumbersome. Yet even with this morphological leverage, word-learning is a
protracted endeavor. To appreciate the task of learning a vocabulary in one's native language,
several factors should be considered: the volume of words that must be learned, the networks of
lexical relationships among words, and the opportunities to learn complex words.
The data about morphology knowledge show that children use many sources of
information from both speech and reading, and that they need extensive experience with
language to fully grasp English morphology. The use of specific verbs and the production of
particular syntactic constructions can be shown to coincide with the frequency of the input
(Kidd, Lieven, & Tomasello, 2006). This type of usage-based approach would account for the
progression of morphology learning, because growth in morphology knowledge can be seen to
parallel the frequency of the morphemes. For example, the ability of elementary school students
to analyze prefixed words is related to the frequency of the prefixes (White, Power, & White,
1989). As research on morphology skill gains more attention, the theoretical background should
be examined more extensively, but one conclusion seems clear: children need ample input for
acquiring morphological word knowledge.
Exposure to Words: The Language Environment
The volume of language and the type of words to which a child is exposed will shape the
child's vocabulary. The number of words can vary tremendously, as was shown in a longitudinal
study of 42 children and their families (Hart & Risley, 2003). The total number of words and the
number of different words that children heard varied with the families' SES levels. Children in
higher income homes both heard and spoke more words than children in low income families.
Over the age range of 13 to 36 months, the children of high income families heard an average of
2,176 words (total) per hour, and 382 different words per hour. In families living at the poverty
level, the children heard an average of 974 words per hour, and about 167 different words per
hour. By the age of 3, the high SES children had vocabularies of about 1100 words, and the
children from low SES homes had about 500 words in their vocabularies.
Even in families with well-educated parents, very frequent words predominate in typical
parent-child conversation (Hayes & Ahrens, 1988) and rare words are most likely to crop up
during book reading or during mealtime conversation (Beals, 1997; Beals & Tabors, 1995). For
example, a study that analyzed the conversations of mothers and 5-year-olds found that less than
1% of the words were "sophisticated" words (outside of the most frequent 3,000 words), and
some children did not hear any less-frequent words in the conversations (Weizman & Snow,
By the time children enter school, differences in their vocabularies can be large. There is
typically little vocabulary gain in the primary grades because of the emphasis on learning to read.
Children are learning to recognize words in print that they already know, so few new words are
introduced (Chall & Jacobs, 2003; Becker, 1977). By the end of second grade, children in the
upper quartile of vocabulary skill may know an average of 8,000 root words, and children in the
lowest quartile average 4,000 root words (Biemiller & Boote, 2006) Schooling in the primary
years seems to do little to augment vocabulary (Christian, Morrison, Frazier, & Massetti, 2000),
which means that vocabulary gaps would persist.
The Distribution of Words: Word Frequency
The American Heritage Word Frequency Book (or WFB) examined the range of words
used in school materials for grades 3 through 9 (Carroll, Davis, & Richman, 1971), and was
based on a corpus of over 5 million word tokens. The word distribution data in this volume put
the word-learning endeavor in perspective. The most common 1,000 word types comprised 74%
of the word tokens, and the first 5,000 types made up almost 90% of the corpus. This makes
learning the English language seem easy: knowing 5,000 words should cover 90% of the printed
material that would be encountered through the ninth grade. On the other hand, about 40% of the
word types turned up only one time in the total corpus of 5 million words. This means that
beyond the first 5,000 most common words, the opportunities to learn specific words may be
Nagy and Anderson (1984) examined the WFB words to better describe the vocabulary-
learning task, and in part to assess whether this level of vocabulary could feasibly be taught.
They selected a random sample of over 7,260 words in the form of 121 "chunks" of 60 adjacent
words. The words were grouped into word families consisting of a base word and all of its
inflections and derivations. A judgment of "semantic relatedness" indicated how transparent or
opaque a target word's meaning would be to a child familiar with the base word (p.310). For
example, an English speaker who knows crowd would readily understand crowded, but the link
between vice and vicious might not be as obvious. Based on their categorizations of relatedness,
Nagy and Anderson added the number of basic words (about 45,000), and the semantically
opaque words (over 43,000) to conclude that printed school English contains 88,533 distinct
word families. Roughly 139,000 words ought to be transparent, so that no additional effort is
needed to understand them. By adding in 'essential' proper nouns, the number of words a student
would need to learn came to about 188,000. Nagy and Anderson concluded that words must be
extensively learned through reading, because it is inconceivable that this volume of vocabulary
could be explicitly taught. Their analysis also showed that basic words tend to be among the
more frequent words, and the derivationally-complex words make up more of the infrequent
Even with these analyses, it is not clear how many words a child needs to learn, but by
most vocabulary methods, the reckoning would be at least tens of thousands. Many of the words
are related, and by understanding the word formation processes of English, learning one basic
word might increase the total vocabulary by a count of several words. In actuality, a word's
meaning may not be completely understood from one encounter, but a reasonable approximation
of meaning might be gained by parsing a complex word into its parts. The extent to which a child
or older reader can 'extrapolate' from basic words to related complex words has not been well-
documented, and it is this word analysis skill that is the main concern of this paper.
Where the Big Words Are: Why Reading Experience is Necessary
According to one estimate, by the time children enter school, most have a basic vocabulary
of the 5,000 or so most common words, learned through conversation and by watching television
(Hayes, 1988). Additional lexical knowledge is contingent on "contextually rich settings" (p.
584), and exposure to words in print.
Children typically master inflectional word changes early in life (Anglin, 1993; Kuo &
Anderson, 2006), but the derivational manipulations of morphemes are much more varied and
require more extensive exposure to language. Although learning about morphemes begins with
spoken language, everyday conversation is unlikely to provide a thorough grounding in
Rare words-defined as those occurring less than 3 times per million words-are truly rare
in conversation and on television, and exposure to these less common words occurs largely
through reading (Hayes, 1988). It is the low-frequency words that tend to be morphologically
complex, and transparent derived words are more common in advanced text than in low-level
material or everyday conversation (Nagy & Anderson, 1984). This means that a considerable
amount of reading is needed for exposure to the less-frequent, complex words and to develop an
understanding of morphology.
When children learn to read, they are initially learning to recognize the orthographic forms
of words they already know (Adams, 1990). Around the fourth grade, the purpose of classroom
materials changes from learning to read to reading to learn (Chall & Jacobs, 2003). The higher-
level reading matter encompasses a broader range of vocabulary with more low-frequency
words, and the proportion of derived, morphologically-complex words increases. It is usually the
difficult, multimorphemic words that communicate the key concepts (Cunningham, 1998; Hayes,
Wolfer, & Wolfe, 1996). This steep change in reading content can lead to a "fourth-grade slump"
(Chall & Jacobs, 2003, p.14) in which children who managed beginning reading skills are
defeated by the vocabulary load. At the same time, good readers have an advantage because
much of the unfamiliar vocabulary consists of derived words. As much as 60% of the unknown
words can be deciphered by analyzing the morphemes (Nagy & Anderson, 1984).
In addition to using longer, more complex words, the nature of academic vocabulary is
different from the vocabulary used outside of school and research settings (Corson, 1997).
Stemming from the history of the British Isles, present-day English language reflects successive
incursions of people bringing different languages. As a result, many ideas can be expressed in
English by words of Germanic origins (primarily Anglo-Saxon) and by words tied to Romance
languages (including Latin and Greek). Everyday conversation favors the Anglo-Saxon words,
while science textbooks and technical materials tend to use words of Latin and Greek origins.
For instance, casual conversation might include the Anglo-Saxon words belly,jaw, or skull, but
in science writing, the Latin terms abdomen, mandible, and cranium would be more common
(Maylath, 1997). With its reliance on Greco-Latin word roots, the formal language used in
advanced academic or research writing has been likened to a second language for English
speakers who lack extensive academic reading experience (Corson, 1997). The notion of
academic vocabulary is difficult to define precisely, but it includes the sub-technical or semi-
technical words that are used in many academic fields, such as phase, inherent, establish, and
analyze (Nation, 2001). This category makes up about 8.5 to 10% of typical academic text
(Chung & Nation, 2003; Nation, 2001; Coxhead, 2000), so even within this genre, the important
words are not high in frequency.
In the early stages of reading, morphology skill may not be essential, but by late
elementary or middle school, knowledge of morphology takes on increasing importance in
reading comprehension (Nagy, Berninger, & Abbot, 2006). This situation can be a boon for avid
readers and a double whammy for struggling readers. Compared to speech, academic reading
material both provides a more intensive exposure to morphology and requires higher levels of
skill. There appears to be a bi-directional relationship in which reading contributes to
morphological skill, and morphological skill enables reading (Nunes, Bryant, & Bindman, 2006;
Nagy, Berninger, Abbott, Vaughan, & Vermeulen, 2003; Voerhoeven & Carlisle, 2006).
Time Spent Reading
In the nation's "Reading Report Card," fourth-grade children described the frequency of
their reading for pleasure (Donahue, et al., 2001). Forty-three percent reported reading every
day; 32% indicated 1 to 2 times per week, and 26% reported that voluntary reading occurred less
than twice a month. These were self-report data, and there is no information about the amount of
time spent reading. Another measure of reading time involved children aged 3 to 11 who
reported their daily activities by means of a time-diary survey (Bianchi & Robinson, 1997).
Trained interviewers asked the children (or parents of younger children) to account for their time
for the previous 24 hours. Only 27% reported doing any reading on the previous day, but 89%
watched television. Across the entire sample of children (n = 887), the average time spent
reading was 12 minutes per day. Some of the children were too young to read independently, so
the data include time in which someone read to the child.
In another study, 155 fifth graders were asked to complete logs to record their activities
outside of school (Anderson, Wilson, & Fielding, 1988). The mean time spent reading books was
about 10 minutes per day, but with all reading combined (books, comics, mail, newspapers, and
magazines) the mean reached about 18 minutes per day. The mean reading times may not be
informative, because the range was "staggering" (p. 294). The median value for total time spent
reading was 7.2 minutes. Students reading at the 90th percentile on a standardized test read 40
minutes per day, while students at the 10th percentile of reading ability typically read 1.6
minutes per day. By multiplying time spent reading by estimated reading speed, Anderson et al.
calculated the average number of words read per year. The spread for the 10th to 90th percentile
students is enormous: 8,000 to 1,823,000 words read per year.
A study of time usage also found that reading did not occupy much of a typical child's day
(Hofferth & Sandberg, 2001). Time diaries were collected for over 2,800 children between 3 and
12 years old. For the entire sample, reading for pleasure took up about 1 hour per week, and
television viewing occupied 12 hours per week. In a longitudinal study of children from first to
fourth grade, a gap in time spent reading at home emerged between good and poor readers in
second grade (Juel, 1988). By fourth grade, the skilled readers reported reading about 3.5 times a
week, compared to once a week for the children with reading difficulties.
The amount of time spent reading in school appears roughly equal to the time spent
reading outside of school. Many school districts subscribe to a trade book program that offers a
wide choice of books and computerized evaluations of students' reading accomplishments. In a
survey of schools using the Accelerated Reader (AR) program, the amount of reading per day
was assessed using the points awarded for each completed AR book, providing a relative, rather
than actual, estimate of reading time (Topping and Paul, 1999). Very few schools allotted the
recommended 60 minutes per day to reading. Private schools allocated about 67% more time to
independent reading than public schools, and small schools reported more reading time than
large schools. Higher reading skill was associated with more time spent reading, but it was
unclear whether practice improved reading, or better readers chose to read more, or both.
Consistent with other studies, there was an enormous difference in the amount of reading done
by skilled and less able readers; readers in the top 5% read 144 times more than those in the
Commercial programs supplying trade books to the schools offer popular, entertaining
books that students self-select. These are very different from textbooks, and some researchers are
concerned about the ability to read and learn from informational text. Familiarity with the genre
of instructional text is important if students are to utilize printed resources in school (Duke,
2000). In a survey of 20 first-grade classrooms, Duke found that very few informational books
were available and they were rarely used. Teachers seldom read aloud from these and students
were unlikely to choose them for independent reading. On average, for all of the classrooms, the
informational books were used for 3.6 minutes per day. The rate was higher in high-SES schools
(3.8% of the time) than in low-SES schools (1.9% of the time).
These studies confirm that the good readers tend to spend more time reading, and that poor
readers generally do very little reading. While any reading is likely to be beneficial, some
reading materials do more to promote vocabulary gain than others. Without adequate exposure to
the more complex vocabulary and syntax of academic or informational texts, it is unlikely that
students will be prepared when they are expected to read textbooks on their own.
Morphology and Spelling
A knowledge of morphology is also important in spelling English words. While English is
an alphabetic language with letters representing sounds, it is not purely alphabetic. Many words
retain spelling that represents morphemes rather than phonology, such as island, \ei' ilu. meant,
and autumn. Young children may start with a phonological spelling strategy, but with age and
increasing experience with writing, they gradually shift to a morphemic approach to spelling
(Nunes, Bryant, & Bindman, 1997; Kemp, 2006). As examples, in the words passed, called, and
waited, the past tense ending has the sounds /t/, /d/, and /id/, respectively. Learning to add the
morpheme '-ed' on the end of regular past-tense verbs is a more successful strategy than trying
to spell the sounds (Nunes, Bryant, & Bindman, 2006).
Measuring Morphology Skill
Previous Studies of Morphology Skill
Children's knowledge of English morphology has been examined with an assortment of
experimental tasks. The earliest studies examined morphological understanding in a global sense,
with more precise differentiation of types of morphology skill appearing in more recent studies.
Many of the newer studies have focused on the acquisition of particular grammatical forms,
using morphology to illuminate the development of syntax. Fewer studies have looked at
morphology as a key to semantic understanding.
In a classic test of morphological knowledge, Berko (1958) tested participants ranging
from young children to adults. By using nonsense words (such as wug), she could eliminate the
effect of familiarity with actual words and test participants' understanding of inflections.
Children did well with inflectional changes such as plurals, possessives, and verb inflections. To
test knowledge of word derivation processes, Berko asked the children to explain the meanings
of compound words. Although familiar words were used, such as birthday, airplane, and
breakfast, the children did not seem aware of the etymological sources of the derived words.
Derwing investigated derivational knowledge by asking specifically about relationships
between pairs of words (Derwing, 1976; Derwing & Baker, 1979). Participants were asked about
examples such as Does teacher come from teach? Does Friday come from fry? Adults showed
much more awareness of etymological relationships than children, who were more likely to agree
that eerie came from ear. Derwing's studies are significant because they elicited and made
explicit the participants' understanding of related words, knowledge which ordinarily may
seldom be verbalized.
Two studies revealed some clues about knowledge of derivational morphology in middle
school students. In a study comparing average eighth graders to high-ability fifth graders, the
students were asked to write definitions on a vocabulary test consisting of basic and derived
words (Freyd & Baron, 1982). The fifth graders outscored the eighth graders, but the difference
was greater for the derived words than for the basic words. This suggests that the high-ability
students had greater understanding of derivational morphology than the average students. In
another test of morphology skill, fourth, sixth, and eighth graders were taught the meanings of
words on a vocabulary list, but were tested on words that were derivationally related to the
training list (Wysocki & Jenkins, 1987). If students had learned connivance and dote, for
example, they were tested on connive and doting. When asked to define the test words, they
more often repeated the definition of the word learned during training, rather than adapting the
explanation to reflect derivational change. While this indicates either limits to definitional skill
or a lack of attention to morphological changes or both, it confirms recognition of the
relationship between base and derived words.
Anglin (1993) studied vocabulary development by asking six- through ten-year-old
children to define words, and he analyzed their responses. In the process of figuring out
unfamiliar words, the children demonstrated their understanding of morphology. For instance,
few of the children knew the word treelet, but some of the fifth-graders knew the word piglet,
and could define treelet by analogy. Anglin called this process "morphological problem solving"
(p. 5) and showed that this skill could account for a significant part of the vocabulary increase in
the older children.
In another study involving fourth, sixth, and eighth graders, students were asked to supply
derivations in response to base words, and to name base words of derived forms (Carlisle, 1988).
In the derivation task, students were given the example "warm/He chose the jacket for its "
(p. 252), and an example of decomposition would be supplying the word danger in a sentence
after hearing dangerous. The students did better at supplying the base words than derived words,
and older students were more successful than the younger students. The most difficult word
conversions involved phonological changes, in examples such as n ihh n i/'e,
explanation/explain, admission/admit, and combination/combine.
This age group has been a focus of study because the middle grades seem to be a time of
active development of morphological understanding, and the skill takes on more importance in
reading at this level. In a recent study, Carlisle (2000) administered her Test of Morphological
Structure to third and fifth graders, asking for derivatives of base words and base words of
derivatives (as described for her 1988 study). In addition, definition and word reading tasks were
used, and reading comprehension and vocabulary tests were administered. The measures of
morphology accounted for 41 to 53% of the variance in comprehension and vocabulary scores,
with the fifth graders demonstrating higher levels of skill than the third graders. In both grades,
students had more difficulty with complex words that require a shift in pronunciation (such as
production/produce), than with transparent words (without phonological shifts).
In another report of this age group, students in third through sixth grades were tested to
compare both morphological skill and phonological awareness with other reading skills
(Mahony, Singson, & Mann, 2000). In the morphology test, adapted from Derwing's method,
pairs of words were presented and children were asked if the two words were related or not. In
the first experiment, children who were better at word decoding also did better on the
morphology test, and after controlling for vocabulary knowledge, morphology still made an
independent contribution to decoding. The unusual finding was that, overall, correct choices on
the morphology relatedness test appeared to be more closely tied to word frequencies than to the
seeming transparency of word relationships. Most of the children in all grades agreed that beauty
and beautiful are related, as are decide and decision, but less than 20% recognized the
relationship in the word pair allow/allowance. Among the word pairs with a phonological shift,
over 90% of children in all the grades knew that add and addition were related, but only about
half the fifth and sixth graders agreed that atom and atomic were related. The second study
suggested that while phonological awareness played a larger role in reading skill, morphological
understanding also made a significant contribution.
Most studies of morphology have examined students in elementary and middle school. In
one of the few studies to test older students, high school and college students were asked to
interpret specially created complex words (Kaye, Sternberg, & Fonseca, 1987). The test words
were formed from well-known prefixes and frequent base words, but the resulting words were
novel combinations, such as submove and compone. Multiple-choice answers offered definitions
based on the various combinations of affix and base meanings. Even the college students had
trouble picking the correct definitions and their answers did not show a consistent strategy;
sometimes they opted for a definition of the affix, and other times chose the definition of the
base word. In other words, being asked to select definitions of many unfamiliar complex words
seemed to be too formidable a task, even though the word parts were familiar. Thus, in contrast
to studies of younger students, these participants did not routinely use the strategy of defining the
word parts in order to decipher an unfamiliar word. Although these participants were older, the
words on which they were tested were unfamiliar, more numerous, and apparently, more
There have been other investigations of morphology use among older students, but these
studies have looked at specific populations of readers. For example, a sample of deaf college
students showed deficits in solving mathematical word problems, even though the students were
capable of performing the relevant mathematical operations when presented in graphic form
(Kelly & Gaustad, 2007). The difficulty seemed related to poor knowledge of morphology, and
when matched on reading level to middle school hearing readers, the deaf students had lower
scores on a test of derivational morphology (Gaustad & Kelly, 2006). The use of morphological
knowledge by dyslexic readers has been explored, and various studies have yielded inconsistent
results. One group of high-functioning dyslexics (college students) was found to process words
with derivational changes more slowly than normal readers, but this effect was not found for
inflections (Deacon, Parrila, & Kirby, 2006). Another study of college students with poor
reading skills showed that the target students, compared to age-matched and reading-level-
matched control groups, were slower at processing derivational word changes. In other cases (but
with younger students), using morphology seemed to provide a compensatory skill with respect
to phonological processing (Arnbak & Elbro, 2000; Casalis, Cole, & Sopo, 2004). With the
various methods, age groups, and languages tested, we do not yet have a clear picture of
morphological strategies used by dyslexic students and other poor readers.
In spite of its apparent value, derivational morphology receives little instructional time in
formal education (Cunningham, 1998; Moats & Smith, 1992) and little is known about the level
of this skill possessed by adolescents and adults. Higher levels of morphological knowledge have
been linked to better reading ability in the early grades (Carlisle, 2003; Nagy et al., 2003; Fowler
& Liberman, 1995; Singson, Mahony, & Mann, 2000). The published studies leave many
questions about the conditions under which a morphological analysis strategy would be applied
in an encounter with a novel word. By middle school most students seem able to recognize the
shared semantic content of common words, although they may not be skilled at defining words in
terms of the correct part of speech. Although children and adolescents can analyze complex
words for meaning, it is not clear that they regularly do this without prompting. Since learning
about derivational morphology continues through high school (Nagy et al., 2003), and even into
adulthood (Hurry, Nunes, & Bryant, 2005; Carlisle, 1988; Kaye, Sternberg, & Fonseca, 1987), it
would be valuable to measure the degree to which adult readers can use their knowledge of
morphology in understanding complex words.
In addition, while morphology is a valid level of language organization for linguists, its
meaning for readers might better be investigated through studies of the metacognitive form,
called morphological awareness.
Making the Case for Morphological Awareness
The process of learning to read, Nagy and Anderson have argued, "is fundamentally
metalinguistic" (1995, p.5), and metalinguistic skill is the ability to reflect on various aspects of
language. Learning to read depends on a particular form of metalinguistic skill called
phonological awareness (or PA). This explicit awareness of the correspondence between sounds
(phonemes) and letters has been definitively linked to progress in reading acquisition (Bradley &
Bryant, 1983, Adams, 1990). Other metalinguistic skills have been implicated in reading, but
with the focus on learning to read, PA has received the most extensive scrutiny.
Once children have mastered the basics of reading, word recognition becomes more rapid
and automatic, and the reliance on phoneme-to-letter relationships declines. When they
encounter unfamiliar printed words, more advanced readers increasingly rely on morphemes in
preference to phonemes (Verhoeven & Carlisle 2006). Several researchers have suggested that
MA is a type of metalinguistic skill that is critical for efficient reading. Morphological awareness
has been linked to early reading skills (Carlisle, 1995) and to children's spelling (Nunes, Bryant,
& Bindman, 1997), and MA appears to gain in importance as reading difficulty increases in
middle school and beyond (Verhoeven & Carlisle, 2006). Much is still unknown about MA
because this research area is still in its infancy, especially when compared to the body of
research on PA.
Young children can create new words from morphemes, but this does not imply that they
possess explicit morphological knowledge (Clark, 1995). Although most adults who are
proficient in English undoubtedly have considerable knowledge of morphology, for many people
this knowledge remains "unconscious" (Hurry et al., 2005, p. 188). A clear distinction between
morphology skill and MA has not been established in the research literature. One leading
vocabulary researcher, William Nagy, has asserted that since much of morphology skill is tacit,
assessments of morphological knowledge are in effect measurements of MA (Nagy et al., 2003).
Just as PA researchers debate the various types of PA and its cognitive underpinnings (Savage,
Blair, & Rvachew, 2006), the investigation of MA will undoubtedly need further refinement.
A commonly used definition of MA includes "children's conscious awareness of the
morphemic structure of words and their ability to reflect on and manipulate that structure"
(Carlisle, 1995, p. 194). The ability to manipulate morphemes can be demonstrated in young
children, such as those who participated in Berko's Wug test (1958), but it is harder to show that
conscious awareness is required.
There is no standardized format for measuring MA, and a variety of approaches has been
used. Carlisle has used a sentence cloze task in which the child must change a word to fit the
sentence, as in "teach. He was a very good or, "runner. How fast can she ?" (Carlisle,
2000, pp. 187, 188). Some items require the derived form to be produced from the base, and
some ask the child to supply the base, given the derived form. Although this task involves the
manipulation of some morphemes, it emphasizes the ability to add or remove suffixes, and to
utilize irregular word forms, as in 1 idei i, i hh Carlisle has claimed this taps explicit awareness
(Carlisle & Nomanbhoy, 1993). In another MA study, children were asked to choose a nonsense
word with the correct suffix, as in "He has lots of STIP now. He's very [STIPful, STIPy, etc.]"
(Windsor & Hwang, 1999, p. 230). Another study used a task described variously as measuring
"morpheme awareness" and "morpheme sensitivity." Participants were asked to choose the
correct word to complete a sentence, in the form of "They were stopped by the with the
choices tramicize, tramify, tramic, and tramity (Singson, Mahony, & Mann, 2000, p. 248). In a
study of second- through fourth-graders, students were given an analogy task of the form a:b::c:?
which was described as a measure of"morpho-syntactic awareness" (Nunes, Bryant, &
Bindman, 2006, p. 773). Examples were cried:cry::drew:_, and happy:happiness: high: (p.
785). Such experimental tasks may utilize the manipulation of morphemes, but many seem to
draw heavily on a form of grammaticality judgment which might be made implicitly. These
methods emphasize the syntactic properties of affixal morphemes, rather than functional
manipulation of the base or stem morphemes that carry the semantic weight.
Other aspects of MA were assessed in two recent studies. Two types of MA along with
other reading skills were measured in kindergarteners and second graders (McBride-Chang,
Wagner, Muse, Chow, & Shu, 2005). The morpheme identification test asked children to pick
one of two pictures that represented the desired morpheme. For instance, with pictures showing
the color blue and "he blew out some air," the child was asked to point to the picture that showed
the same meaning of blue in blueberries (p. 423). To demonstrate understanding of
morphological structures of words, the children were asked to create compound words to fit a
described situation. As an example of this test, after being given an explanation of the word
sunrise, a child would be asked what to call the moon coming up at night. Correlations of the two
MA tasks with a standardized picture vocabulary test were between .34 and .50. After entering
the PA and other reading scores in the first step of a hierarchical regression, the MA skills
accounted for 10% of the variance in vocabulary for the combined sample of children.
In another study comparing the contributions of PA and MA to reading, fourth through
ninth graders completed a battery of reading tests (Nagy et al., 2006). The MA tests included a
judgment requiring recognition of a grammatically-correct suffix, and a version of Derwing's
'comes from' test, as in 'Does moth come from mother?' Using structural equation modeling,
these forms of MA were found to contribute unique variance to the reading comprehension and
vocabulary at each of the grade levels. The results suggested that MA played an increasingly
important role in the reading skills as grade level increased.
These studies show that MA contributes to reading facility in various ways, and more
specific means of classifying MA skills are needed. Weng, Cheng, and Chen (2006) have
proposed four categories of MA, including the identification of morphemic components in
complex words, the discrimination of morphemes from similar-sounding words, and the
manipulation of morphemes to produce syntactically-correct words. The fourth proposed
category, interpretation, refers to the ability to infer the appropriate meaning of complex words
from their morphemes, and this is the skill of most interest in this study. The ability to extract
meaning from word parts would be valuable for reading advanced level text, and it has rarely
been tested using college-level vocabulary and readers beyond middle school.
A few studies have shed light on this semantic or interpretative form of MA by
interviewing participants and asking them to explain how they figured out word meanings
(Anglin, 1993; Baumann, Edwards, Font, Tereshinski, Kame'enui, & Olejnki, 2002; Freyd &
Baron, 1982). This method provides a detailed, step-by-step rendition of the processes children
use to decipher unknown complex words. While these introspective reports provide useful
information, such time-consuming procedures limit the number of participants from whom data
can be collected and the number of words that can be tested. One goal of the present study was to
develop an instrument that could be administered to a group of participants in a relatively brief
While many researchers agree that MA is important, there is no consensus on how to
categorize it or how to measure it, or even how to tease it apart from a basic-level, non-
metacognitive morphological knowledge. The existing studies are largely confined to young
readers, so little is known about the influence of MA on more advanced reading. Although the
value of MA in understanding complex words is acknowledged, there is little empirical evidence
of this application of MA. In response, the main goal of this study was to measure MA as it
contributes to semantic understanding, to test a measure appropriate for adult readers, and to
make sure the task utilizes explicit word knowledge.
The Rationale for the Morphological Awareness Measure
In this study, MA was assessed through a roundabout, two-step route. The method is based
on the assumption that a person who knows a complex, derived word should know the base word
which is more frequent, and therefore, normally learned earlier. For example, someone who
knows disagreement should also know the word agree. The word agree would be expected to be
learned within the first 2,000 words of English, whereas disagreement is ranked at about the
14,000-word level (Carroll, Davies, & Richman, 1971). It is hard to imagine that the relationship
between words such as agree and disagreement might not be recognized, but both are fairly
frequent words. The derivational relationships of less common complex words and their bases
may be less obvious, such as in beneficiary and benefit or judicial and judge. Recognition of
derived words appears to involve awareness of the word family (Marslen-Wilson, Tyler,
Waksler, & Older, 1994; McKinnon, Allen, & Osterhout, 2003; Nagy, Anderson, Schommer,
Scott, & Stallman, 1989), so claiming knowledge of a complex word implies the understanding
of its derivational relationship to the basic word.
In the first step of this study, the participants were presented with a list of 50 complex
words, and were asked to check the words that they know. It was assumed that participants
would check words for which they knew the general meaning.
The second step presented the same list of 50 words, and asked the participants to identify
a base word to which each complex word is related. Sample items include
believable a. belief b. belie c. lie
discredited a. disc b. credit c. edited
explanatory a. plane b. planetary c. explain
By itself, the second step could be seen as a measure of basic-level morphological skill,
and for this, knowledge of the particular words would not be necessary. To extricate MA from
this procedure, the results of the second step were examined with respect to words checked as
known in the first step. In other words, to demonstrate MA, the participants needed to indicate
that they knew specific words and to show they knew the morphologically related words. This
two-stage sequence makes explicit the knowledge of morphological relationships that may
normally be used implicitly.
The multiple choice format for the morphology task allows a higher level of difficulty so
that the measure is appropriate for adults. The results of the newly-created morphology test were
compared to the participants' scores on a standardized test, the Nelson-Denny Reading Test
(Brown, Fishco, & Hanna, 1993). It was expected that results on the MA test would be positively
correlated with the vocabulary and reading comprehension tests, in support of Anglin's view
(1993) that morphological skill contributes to a large portion of vocabulary growth.
As a third step, the participants were also given a multiple-choice vocabulary test using the
same list of complex words. This allowed the testing of another type of metalinguistic skill, the
ability to recognize whether or not a word is known. This form of 'knowing what you know' was
In summary, the test designed for this study asked three questions for each of 50 complex
words: (1) Do you know the word? (2) Do you know the morphologically simpler word? (3) Do
you know the meaning? In addition to knowing its meaning, knowledge of a complex word
should include the word's place in a word family, and measuring this type of MA was the
primary purpose of the test design.
Measuring MA required a means to tap awareness of knowledge that is normally implicit.
Since this ultimate goal depended on the validity of the individual steps in the testing process,
the hypotheses included predictions about separate sections of the morphology test. Most of the
hypotheses involve comparing scores on the 3-part morphology test with scores on the Nelson-
Denny Reading Test (ND). The first two hypotheses refer to part two of the morphology test
which measures basic morphology skill (without measuring awareness of this knowledge):
1. Students with higher ND vocabulary scores would have better morphology skill than
those with weaker vocabularies.
2. Students with better ND reading comprehension would have better knowledge of
morphology than students with lower comprehension scores.
Hypotheses 3 and 4 refer to the measure of MA that is derived from the first and second
3. Students with higher ND vocabulary scores would have higher MA scores.
4. Students with higher ND reading comprehension scores would have higher MA scores.
The fifth hypothesis addresses the issue of transparency in word relationships:
5. On the morphology test, scores would be higher for complex words that retain the base
word without changes of spelling or phonology (or with very minor changes).
The final hypothesis involves accuracy, another metalinguistic skill. It was predicted that:
6. Students with higher ND vocabulary scores would be more accurate at gauging their
knowledge of word meanings than students with lower vocabulary scores.
Undergraduate students in psychology courses at the University of Florida were recruited
for the study, and participation contributed to the fulfillment of course requirements. There were
originally 108 participants, but 2 people were dropped from the analyses due to missing data.
(One had to leave the test early and one skipped a page of the test.) Of the remaining 106
participants, there were 25 males and 81 females, and most were between 19 and 21 years old (M
= 20.42; SD = 1.64). The majority reported that they spoke and read English fluently, with
80.2% speaking only English, and 84.0% reading only English. (Table 3-1). Of the other
language use categories, the next highest percentages were for those who spoke both English and
Spanish fluently (12.3%), and those who read fluently in both English and Spanish (8.5%).
A Standardized Reading Test: The Nelson-Denny Reading Test
The Nelson-Denny Reading Test (ND) was selected as a standardized test because it is
suitable for group administration and it provides norms for college students. The ND is a well-
regarded instrument used both for research purposes and for assessment of reading progress in
schools and in clinical settings (Masterson & Hayes, 2004), and Form G was used in this study.
Standardization of the test was done from 1991 to 1992, and the college-level sample included
over 5,000 students from 38 four-year universities (Brown, Fishco, & Hannah, 1993). Standard
administration includes a 15-minute vocabulary section with 80 items, and a 20-minute
comprehension section with 38 questions. Each multiple-choice question in the vocabulary and
comprehension sections has five answer choices. As part of the comprehension test, at the end of
the first minute, participants are instructed to mark the number at the end of the line they are
reading, and this provides the reading rate. The ND provides scores for vocabulary and reading
comprehension, plus a combined score and the measure of reading rate.
The Morphology Test Designed for the Current Study
The test for measuring morphology skill contained three sections, which were called
Subtest 1: Checklist, Subtest 2: Morphology Skill, and Subtest 3: Definitions. Each section
required a response to the same list of 50 morphologically complex words. (Appendix A.)
The items on the list ranged from fairly common words (apparently, dependable, and expensive)
to words of low frequency (impiety, indefatigable, and redacting). Some of the words did not
appear even one time in the WFB (Carroll, Davies, & Richman, 1971), so it can be assumed that
their frequencies were extremely low. To the degree that word frequencies could be determined
in the WFB, complex words were chosen that are much less frequent than the simpler words. (In
English, there are a few complex words that occur more frequently than their base words, and
these would not be suitable for this measurement of MA.) Compound words had to be avoided to
prevent ambiguity in selecting a base word. The primary challenge in selecting test words was to
find complex words containing two smaller words that would be plausible answer choices (aside
from the correct answer), as in these examples:
noncombatant a. comb b. bat c. combat
diversification a. diverse b. versify c. diver
For Subtest 1: Checklist, participants were given the list of 50 complex words, and asked
to place a check next to the words they know. The instructions included "check yes if you could
use the word in a sentence." In everyday parlance, knowing a word means recognizing its
meaning, and the superficial purpose of Subtest 1 was simply to obtain a yes-or-no response for
this level of knowing each word. The less obvious purpose was to gain access to a deeper level
of knowledge in later steps of the test. A thorough knowledge of a word includes its associations
with derivationally-related words. This knowledge is usually implicit, but can be made explicit
by the sequence of Subtest 1: Checklist followed by the second subtest.
In Subtest 2: Morphology Skill, the participants were asked to identify the simpler word
that is morphologically related to each of the complex words (as shown in the example above).
By itself, the score for Subtest 2 would represent the student's basic knowledge of general
derivational word-formation processes (or morphology skill). Taking results on Subtest 2 for
only the words marked known on Subtest 1 represents the specific, usually tacit knowledge of the
derivational relationships of the particular test words. This is how the score for morphological
awareness (MA) was obtained.
Subtest 3: Definitions was a standard-format, multiple-choice vocabulary test, in which the
definition of the target word had to be selected. A sample item from the third subtest was
a. lowering of rank
c. without moving
The score for Subtest 3: Definitions is simply a measure of knowledge of word meanings.
Comparing Subtest 3 answers to words marked as known on Subtest 1 can be used as an
assessment of metacognitive skill. This comparison was termed Accuracy because it indicates
whether or not the students actually know the meanings of the words they think they know.
In addition to the three-part morphology test, the participants were also asked about the
amount of reading they do, and if they had any language problems in school or difficulty
learning a foreign language.
Scores on the ND were obtained in accordance with the instructions in the test manual. The
numbers of correct items on the Vocabulary and Comprehension tests were recorded, and these
were subsequently translated to raw scores, and then to standardized scores. Percentile ranks for
each student were also obtained.
For the morphology test, three subtest scores were obtained. Starting with Subtest 1:
Checklist, the number of words checked as known was recorded. Scores for Subtest 2:
Morphology Skill, and Subtest 3: Definitions were the numbers of correct responses on each
subtest. The responses to individual words on the three subtests allowed the scores for MA and
accuracy to be derived. (Figure 3-1.)
The MA score was calculated by counting the correct answers from Subtest 2: Morphology
Skill for words checked as known (on Subtest 1: Checklist). Similarly, Accuracy scores were
obtained by assigning one point for each correct definition from Subtest 3: Definitions for each
word checked as known (on Subtest 1: Checklist). As an example, if a student marked
commendable as unknown on the Checklist, but correctly chose commend as the related word on
Subtest 2: Morphology Skill, the student would be credited with one point toward the Subtest 2
score, but would not earn credit for this word on the MA score. Using the same procedures, if a
student marked opacity as known, but did not select the correct definition on Subtest 3:
Definitions, no point would be credited to the Accuracy score.
Testing was done in groups, and the session took about one hour. After giving informed
consent, the participants provided some demographic information and answered questions about
their language and reading backgrounds. The ND was administered next, followed by the three
subtests. Students had to complete the three sections in order, without returning to previous
sections. Subtest 2: Morphology Skill included a timed measure and all participants started this
section at the same time. After 3 minutes, the students were instructed to circle the number of the
last item completed, and they were then given as much time as needed to complete this section
(usually just a minute or two).
Table 3-1. Languages participants use fluently.
English only 85 (80.2%)
English and Spanish 13 (12.3%)
English and other 6 (5.7%)
Subtest 1: Checklist
2: Morphology skill
Number of correct items
on Subtest 2 for which
words were checked as
known on Subtest 1
Number of correct items
on Subtest 3 for which
words were checked as
known on Subtest 1
Figure 3-1. Scoring of the morphology test.
The Nelson-Denny Vocabulary, Comprehension, and Total Reading scores were compared
by grade level, and the four groups (freshmen through seniors) did not differ significantly. Also,
there were no grade level differences for the morphology test scores, so grade level was not used
in subsequent analyses, except for determining the percentile rank for each student according to
the ND norms. A comparison of the Spanish/English bilingual students and the monolingual
English-speaking students also found no correlations between language fluency and reading test
No gender differences were found in the ND raw scores for Vocabulary, Comprehension,
or Total Reading. The scores for the morphology test did not differ by gender, although the tests
for differences on the MA and Subtest 1: Checklist scores approached significance, with p values
of 0.052 and 0.054. In both of these cases, the average score of the male students was higher than
for the female students. In light of the relatively large number of comparisons, and the lack of
differences on the standardized test, gender was not considered in further analyses.
Amount of Reading
Roughly 76% of the students reported reading less than 4 hours per week for fun, and
about 74% claimed that time spent reading for class was less than 8 hours per week (Table 4-1).
Most students (about 67%) reported that during a year they read 5 or fewer books for fun. Only
17% indicated that they chose to read more than 10 books in a year. Correlations were run
between the amount of reading and the ND and morphology test scores, but no significant
correlations were found.
Participants' Reading Background
In questions about their reading background, a small proportion of students described
learning difficulties that had been identified or for which they received tutoring. These included
speech therapy (7.5%), a reading problem (2.8%), and learning difficulty (2.8%). None of the
students reported any other type of language problems that had been formally identified. Another
set of questions asked about problems they may have noticed, but that were never diagnosed. In
this section, 11 students said they noticed a spelling problem (10.4%), and 5 reported a language
difficulty (4.7%). This line of questioning was intended to pick up additional reading problems,
but there may have been some misunderstanding. Two students indicated both an identified
reading problem, and a reading problem that they noticed but that was not formally recognized.
While it is possible that they might have two separate reading problems, it is more likely these
two answers represent a duplication. Counting these 2 students only once for the identified
problem, there were 8 students (7.5 %) who thought they had reading problems that had not been
diagnosed. Another 10 students reported trouble learning a foreign language. A learning
difficulty in a family member was reported by 15 students, and about half of these relatives were
brothers, but the numbers in each category of family member were too small to make a valid
comparison. About a third of the college students reported having received tutoring for some
subject in the past, and the tutoring subjects very mostly math. Only 3 students (2.8%) were
tutored in reading.
Due to the small number of students with learning problems, the problems were combined.
For each student, one point was added for each affirmative answer to a question about a learning
difficulty or a family member with a learning difficulty. This step resulted in 67 students (63.2%)
reporting no learning problems, 25 students (23.6%) with one instance, and 14 students (13.2%)
with between 2 and 5 various problems. Comparing the total number of learning problems and
gender, the means for the female and the male students did not differ. Correlations were run
between the total problems and the ND and morphology test scores, and no correlations were
Nelson-Denny Reading Test
The ND test yielded raw scores for Vocabulary, Comprehension, Total, and reading rate,
and these scores are shown in Table 4-2. For the Vocabulary and Comprehension tests,
participants from the University of Florida on average, had reading skills higher than those in the
ND standardization sample, particularly among the freshmen, sophomores, and juniors. The
average scores of the seniors were very close to the ND norms. The average reading rates for the
study participants were below the ND norms.
Morphology Test: Subtests 1, 2, and 3
On Subtest 1: Checklist, participants indicated that they knew most of the 50 words. For
Subtest 2: Morphology Skill, and Subtest 3: Definitions, most of the items were answered
correctly. A timed measure on Subtest 2: Morphology Skill provided the number of items
completed in 3 minutes. These scores are shown in Table 4-3.
The derived scores, MA and Accuracy, were calculated from the three subtests, and these
are also shown in Table 4-3. These scores for metacognitive skills show that for words they
claimed to know, students correctly picked the morphologically-related word for about 39 of the
50 words, and identified the definition for about 40 words.
Correlations between the three Subtest scores, the two metalinguistic scores, and the ND
raw scores are shown in Table 4-4, and all were significant (p < .05). The correlations of greatest
interest were those between the main scores of the ND (Vocabulary, Comprehension, and Total),
and the morphology tests devised for this study.
The correlations of Subtest 2: Morphology Skill and the ND Vocabulary and
Comprehension scores were used to test hypotheses 1 and 2. These hypotheses were that:
1. Students with higher ND vocabulary scores would have better morphology skill than
those with weaker vocabularies.
2. Students with better ND reading comprehension would have better knowledge of
morphology than students with lower comprehension scores.
These correlations were positive and significant, but not very large (r =.43 for Morphology and
ND Vocabulary; r = .27 for Morphology and ND Comprehension), but the results did support the
Hypotheses 3 and 4 were
3. Students with higher ND vocabulary scores would have higher MA scores.
4. Students with higher ND reading comprehension scores would have higher MA scores.
The correlation of MA and the ND Vocabulary scores was .64, and the correlation of MA and
ND Comprehension scores was .45, so hypotheses 3 and 4 were supported by the results.
The ND vocabulary scores were compared with the Accuracy scores for the sixth
6. Students with higher ND vocabulary scores would be more accurate at gauging their
knowledge of word meanings than students with lower vocabulary scores.
The result was r =.65 for Accuracy and ND Vocabulary, which supports hypothesis 6. In
addition, a correlation of r = .52 was found for Accuracy and ND Comprehension. (Hypothesis 5
will be discussed below.)
The strongest correlations were between the ND scores and Subtest 3: Definitions, ranging
from .63 to .73. The measures for Subtest 1: Checklist, Subtest 2: Morphology Skill, and the
timed 3-minute score showed weak to modest correlations with the ND scores (.27 < r < .49).
To further investigate the MA scores, the number of errors across all participants was
calculated for each item on Subtest 2: Morphology Skill. The 50 words were split in half, and
assigned to one of two categories: easy words (0 to 3 errors), and hard words (4 to 77 errors).
Correlations were run for the 25 harder items and the ND scores, which produced slightly higher
correlations for ND Comprehension and Total scores, but no change in the correlation with the
ND Vocabulary test. (Table 4-5.)
In a standard linear regression analysis, three variables, MA, Subtest 2: Morphology Skill
score, and Subtest 3: Definitions score, were tested as predictors of the ND Total Reading score.
Combined, the three independent variables accounted for over half (R2 = .543) of the variability
in the ND Total Reading scores, although the separate contributions of the morphology and MA
scores were not significant. The score for Subtest 3: Definitions was the best predictor of the ND
Total Reading score, as shown in Table 4-6.
Hypothesis 5 was related to transparency of word relationships:
5. On the morphology test, scores would be higher for complex words that retain the base
word without changes of spelling or phonology (or with very minor changes).
To test hypothesis 5, the relationship of each complex word to its base word was assessed to
indicate the degree of transparency (or opacity) of the relationship. This was quantified by
counting the number of spelling and phonological changes between the two words. As an
example, the word protestation contains an intact base word, protest, and there are no
phonological changes to the base word inside the complex word, so it was scored as 0 changes.
The changes to the word decide in indecision are more numerous: 2 letters of decide are deleted,
the vowel i changes from a long to short sound, and the consonant after the i changes, so this was
counted as 4 changes. In examining the number of changes and the number of errors for each of
the 50 word pairs, no correlation was found.
Self-assessment of Reading and Vocabulary
The participants were asked to rate their own reading skill and vocabulary knowledge, and
the vast majority reported that their skills are at least adequate for the demands of college-level
reading. Only 3 students reported less-than-adequate vocabulary, and only 4 thought their
reading skills were insufficient. One of these students rated both her vocabulary and reading skill
as low, but the other 6 indicated low skill in only one category. For the 3 claiming low
vocabulary skill, percentile ranks on the ND Vocabulary test were between 6 and 27 percent. For
the remaining 3 who rated their reading skills as low, only one seemed to be accurate, with a
percentile rank lower than 10%. The other two students had percentile ranks between 40 and
80% on the ND tests. Using the ND percentile ranks, the number of students in each quartile was
examined, as shown in Table 4-7. For the Vocabulary, Comprehension, and Total Reading
scores, all but 16 to 18 students (15 18%) were above the lowest quartile, confirming that the
level of reading skill for most students should be adequate.
Comparison of Better and Less-skilled Readers
Using the percentile ranks of the ND Total Reading scores, participants were divided into
two groups of reading ability, those in the lower half and those ranked at 51% or above. Using an
ANOVA, the means for the two groups were then compared on four aspects of the morphology
test: Subtest 2: Morphology Skill, Subtest 3: Test of Definitions, the timed measure of
morphology skill, and MA. (Table 4-8.) For all 4 measures, the means of the two groups
were significantly different with higher means for the group of better readers. On all four tests,
the difference in means between groups was significant (p < .01).
Table 4-1. Amount of reading per week during the school semester.
Purpose of reading
For fun For class
Number (Percent) Number (Percent)
Less than 2 hours 54 (51%) 8 (8%)
More than 2 hours, but less than 4 hours 26 (25%) 32 (30%)
More than 4 hours, but less than 6 hours 9 (9%) 25 (24%)
More than 6 hours, but less than 8 hours 8 (8%) 13 (12%)
More than 8 hours, but less than 10 hours 3 (3%) 7 (6%)
More than 10 hours 5 (5%) 21 (20%)
Table 4-2. Nelson-Denny Reading Test scores for the standardization sample and University of
Standardization sample University of Florida
Mean SD Mean SD
Vocabulary 52.43 15.12 55.33 7.37
Comprehension 51.73 15.19 59.33 14.19
Total 103.45 28.78 114.67 21.50
Reading rate 238.46 87.03 207.33 21.36
Vocabulary 58.70 14.30 60.09 12.46
Comprehension 55.57 14.59 58.82 14.30
Total 113.94 27.03 117.55 25.63
Reading rate 246.81 85.54 218.05 50.78
Vocabulary 61.21 13.47 65.86 9.62
Comprehension 57.28 14.15 62.34 11.45
Total 118.72 25.74 128.20 19.28
Reading rate 251.91 83.77 242.59 68.86
Vocabulary 64.52 11.46 64.89 9.68
Comprehension 61.60 11.94 62.52 9.23
Total 126.56 21.91 126.98 17.12
Reading rate 257.65 83.03 240.75 78.08
Note: There were 106 participants from the University of Florida (3 freshmen, 22 sophomores,
35 juniors, and 46 seniors). The Nelson-Denny standardization sample included 1,043 freshmen,
488 sophomores, 584 juniors, and 558 seniors. The Nelson-Denny data are from Brown, J. I.,
Fishco, V. V., & Hanna, G. S. (1993). Nelson-Denny Reading Test: Technical Report Forms G &
H. Itasca, IL: Riverside Publishing.
Table 4-3. Performance on morphology test subtests and metalinguistic measures (n = 106).
Mean SD Rangea
Subtest 1: Checklist 42.14 4.34 26-49
Subtest 2: Morphology skill 45.05 4.18 23 50
Subtest 3: Definitions 45.73 3.70 33 -50
3-minute score, Subtest 2 43.66 7.76 18 50
MA (Morphological awareness) 38.99 5.11 24 49
Accuracy 39.92 5.13 22-49
maximum possible score = 50
Table 4-4. Intercorrelations among measures.
1 2 3 4 5 6 7 8 9
MA Accuracy Subtest 1 Subtest 2 Subtest 3 3-minute ND ND ND
checklist morphology definitions timed vocabulary comprehension total
Note: *p < 0.05; **p < 0.01
Table 4-5. Correlations of Nelson-Denny and MA (Total and hard words) scores.
Nelson-Denny Reading Test
Vocabulary Comprehension Total
MA Total (50 words) .642 .448 .603
MA Hard words (25 words) .642 .496 .634
All were significant at the p < .01 level.
Table 4-6. Linear regression predicting Nelson-Denny Total Reading (raw scores) from three
Variable Beta t-value
Subtest 3: Definitions .584 6.04*
Subtest 2: Morphology skill .022 0.22
MA .182 1.71
*p < 0.001
Table 4-7. Performance of good and less-able readers
Score Mean scores by group F p value
1 -50 51-100 statistic
Subtest 2: Morphology skill 43.78 45.98 7.68 .007
Subtest 3: Definitions 43.82 47.13 25.52 .000
3-minute Morphology measure 39.93 46.41 21.59 .000
MA 37.11 40.36 11.66 .001
Table 4-8. Numbers and percentages of participants in each quartile of Nelson-Denny scores.
First Second Third Fourth
No. (Percent) No. (Percent) No. (Percent) No. (Percent)
Vocabulary 16(15.1%) 27 (25.5%) 38 (35.8%) 25 (23.6%)
Comprehnsn. 18(17.0%) 28 (26.4%) 30 (28.3%) 30 (28.3%)
Total reading 16(15.1%) 29 (27.4%) 33 (31.1%) 28 (26.4%)
Reading rate 37 (34.9%) 28 (26.4%) 27 (25.5%) 14 (13.2%)
Findings of the Current Study
The goals of the current study were to develop a measure of MA in young adult readers and
to examine its association with reading ability, and in particular, vocabulary skill. While there are
many strategies children and adults use to acquire new vocabulary, morphological knowledge may
be particularly important for later vocabulary development as older children and adults encounter
more complex reading material. There are many good reasons for expecting MA to be linked to
reading ability in adults, but as yet, there is little empirical evidence. There has been only limited
testing of MA in adults, especially in conjunction with standardized measures of reading skill, and
this study represents an initial foray into this area. Overall, the results showed that better readers
and those with strong vocabularies had better metalinguistic awareness than less-adept readers.
The new method shows promise and deserves additional development.
The conceptual underpinning of this study assumes that there is a bidirectional relationship
between MA and reading. The evidence that reading promotes MA is based largely on
characteristics of the English language. The scope of words used in writing is much broader than
the vocabulary of everyday conversation. Some understanding of morphology is gained through
spoken language, but exposure to morphemes would be limited without extensive reading. It
appears that reading would be essential to gaining MA in English, and by default, this leads to the
corollary that reading must be the impetus that spurs morphological understanding.
The reasoning behind the second element-that MA promotes reading--is this: vocabulary
knowledge contributes to reading comprehension, and morphology knowledge contributes to
vocabulary. Also, awareness of one's knowledge (explicit knowledge) represents a deeper
understanding than implicit skill. Therefore, an explicit knowledge of word formation principles,
and the clues to meaning that result, should contribute to skilled reading.
These claims seem plausible, but the separate steps have not been fully tested. Vocabulary
knowledge is closely tied to reading comprehension, and correlations between vocabulary and
comprehension tests are often in the .80 range (Bloom, 1976, cited in Biemiller & Boote, 2006).
Although reading comprehension requires more than vocabulary, knowledge of the words is
certainly a major factor. As for the next step, the relationship of morphology to vocabulary has
some empirical support, but this has not been widely investigated, and the studies are limited
mostly to elementary and middle school students. Finally, although several studies have claimed to
measure the specific awareness of morphology, the distinction between implicit and explicit
knowledge is still fuzzy. The developmental progression of morphology skill has yet to be fully
analyzed and described.
Two lines of reasoning suggest that MA should increase through the teenage years and into
early adulthood. First, previous studies show that morphology knowledge is growing in middle
school, but is far from complete. In addition, as students advance in school, the level of the
reading material increases in difficulty. These two conditions set the stage for MA to increase
through high school and early adulthood, (assuming that sufficient reading is done). The next
logical step, and the primary goal of this study, was to study the development of MA in older
students and adults by developing a new method of assessing MA.
To evaluate the new method, the various elements of the morphology test were compared to
the results on the Nelson-Denny (ND), a standardized reading test. The first two hypotheses were
that the scores on tests of Morphology Skill (Subtest 2) would be correlated with the ND
Vocabulary and Comprehension scores. The results supported these hypotheses, although the
correlations were modest. The Morphology Skill scores were more highly correlated with the ND
Vocabulary scores than with the Comprehension scores, as would be predicted from the role that
morphology is thought to play in reading. These results also reflect the fact that reading
comprehension requires an array of skills, and not just word knowledge.
The next two hypotheses stated the key predictions of the study-that the MA measure
would be related to the ND Vocabulary and Comprehension scores, and these hypotheses were
confirmed by fairly strong correlations. Again, the correlation of the experimental measure for
MA was more strongly related to Vocabulary than to Comprehension. Also, the correlations
involving the derived MA score were higher than those using the basic Morphology Skill score,
which lends support to the concept of measuring explicit knowledge of morphology as opposed to
a basic, implicit level of morphology skill. The data suggest that this test of MA taps an ability
that is closely related to the knowledge of word meanings.
The fifth hypothesis was based on results from previous studies, and it predicted that
transparent word relationships would show higher scores on Morphology Skill (Subtest 2). In most
studies of MA, the degree of transparency of word relationships has influenced the MA scores,
because phonological and spelling changes can mask the relationship between base word and
derived forms or make the similarities readily apparent. For example, phonology has an influence
in that young readers are more likely to recognize the relationship of culture to cultural than of
nature to natural. Fewer changes make the relationship of happy to happiness easier to see than
the relationship of pronounce to pronunciation, but the influence of phonological changes
compared to spelling changes is difficult to assess. Previous studies have evaluated transparency
in a variety of ways, and there is no clear-cut method for rating the transparency or opacity of
word relationships. For the adult readers in this study, however, the transparency of the word
relationships did not seem to be a factor. This is likely due to the familiarity of the words to the
participants, but it would also reflect the higher levels of MA possessed by these older readers in
comparison to studies of younger students.
The final hypothesis was that students with better vocabulary skill, as measured by the ND,
would also be more accurate at assessing their knowledge of individual words, and results
supported this hypothesis. The validity of self-ratings of word knowledge (or the ratings of parents
of their children's vocabulary) has been a methodological issue in language research, but this type
of measure also has applications for instructional design and curriculum planning. If students are
wrong in thinking their word knowledge is sufficient, they may not make an effort to improve.
Evaluating this metacognitive aspect of vocabulary would be useful due to its implications for
reading and vocabulary instruction.
Correlations were also examined between the scores on Subtest 3: Definitions and the ND
scores. The third subtest was a basic test of knowing the meanings of words, and it used a format
that is typical for a vocabulary test. The relatively strong correlations may reflect the greater
familiarity of a multiple-choice vocabulary exercise, better-designed test items, or some
combination of factors. If the element of the test measuring basic vocabulary skill had not been
positively related to the ND Vocabulary scores, the value of the test would be seriously
undermined. The large correlation between the ND Vocabulary test and Subtest 3: Definitions
affirms that this aspect of the morphology test is at least 'in the ballpark.'
While five of the six hypotheses were supported, there were also some unexpected results in
the study. It was predicted that the reported amounts of reading would be positively correlated
with the scores on the ND and on the morphology test, but this was not the case. Surprisingly, no
relationships between amounts of reading and test scores were found. It is possible that the
question about time spent reading for class was interpreted to mean literally reading a textbook
and that this did not include reading notes or other studying. It is also possible that reading skill
was developed prior to college entry, and that current reading habits are not related to reading
ability, although the norms for the ND test do not fully bear out this idea because they show
increases in vocabulary and comprehension scores during the four years of college. More detailed
questions about reading habits may have been needed.
The biggest limitation of the study is related to its level of difficulty, because the test turned
out to be relatively easy for most of the participants. As a group, the students who participated
were better-than-average readers. Their ND Vocabulary, Comprehension, and Total scores were
above the norms for each year of education. The fact that these participants were almost uniformly
good readers probably contributed to a ceiling effect in the results of this study. For a sample of
highly competent readers, it would have been better to include more difficult test items. Of the 50
complex words tested in this study, most were familiar to the participants, with an average of 42
words checked as known. The third subtest showed that the participants could pick the correct
definition for almost 46 of the 50 words, so very few of the items were unfamiliar. More obscure
and complex words could be included in future versions of the test, although there is little
information available about levels of difficulty in this type of word knowledge.
In questions about their reading and education history, few of the participants noted that they
had ever experienced problems in reading. This is another indication of the high level of relevant
skills in this sample. The incidence of reading disability or dyslexia in the overall population has
been reported as between 5 and 17.5% (Shaywitz & Shaywitz, 2003; Katusic, Colligan, Barbaresi,
& Schaid, 2001), depending on the definition and the diagnostic criteria used. Of the college
students who participated in the present study, only 3 (2.8%) reported having received tutoring in
reading, although 9 (7.5%) had received speech therapy and more suspected that they might have
had some difficulties in learning language. It is possible that students who have experienced
reading difficulties might be less likely to be admitted to a top-ranked state university than
students without reading problems. No correlation was found between the history of reading
problems or learning disabilities and the ND scores, and this was probably due to the sample size
in this study. If an average of 10% of a population is dyslexic, then a much larger sample should
be used to investigate the morphological skills of students with reading difficulties.
Based on these results, the mechanics of the morphology test seem sound, but a wider scope
of test items is needed. Including more obscure words could extend the upper range of the test.
The multiple-choice format can create a more challenging test than the 'comes from' test, and the
main problem is finding suitable items to test. The current test used only free stems (whole words)
as answer choices, but using more distantly-related members of word families could increase the
difficulty. Recognizing words with bound stems in common (such as transmittal and admission, or
scribble and nun'ie / ip') is more challenging than noticing shared free stems (as in readmission
and admit). Such items could measure a more abstract level of morphological skill.
Reasons to Pursue Morphological Awareness Research
In spite of the high stakes involved in teaching children to read, MA has been almost
overlooked by many reading researchers. With the increasing attention being given to reading
problems of older students, morphological awareness (MA) warrants closer investigation. Even
though a deficiency in morphology skill may not be apparent in young students, the full
developmental course of morphological knowledge needs more study.
Several studies have examined both PA and MA in young readers, and PA has retained its
crown as the key element in learning to read, with MA making a much smaller contribution
(McBride-Chang, Wagner, Muse, Chow, & Shu, 2005; Nagy et al., 2006; Mahony, Singson, &
Mann, 2000). Still, it is difficult to completely disentangle PA from MA, and these skills have a
fair amount of overlap (McBride-Chang, et al., 2005). For example, the classic elision test leads
off with easier questions such as 'Can you say popcorn without the pop?' The pop can be labeled
a phonemic unit at the level of the syllable, but it is also a morpheme. It may be that MA helps
children to isolate phonemes in the speech stream, and that early morphological skill has been
subsumed by some PA tasks. Although in learning to read, the critical nature of PA seems
unequivocal, these results do not mean that MA is unimportant for more advanced reading.
Without well-defined tasks for measuring MA and tasks that are appropriate for older readers, it is
difficult to completely disentangle the PA and MA components of reading and to organize them in
a chronological pattern of development.
While explicit awareness of morphology is presumed to be valuable, this assumption seems
based on an analogy to PA, rather than on empirical evidence. Interest in MA is growing, but
research in this area is still relatively sparse. Some organizational frameworks for morphology and
MA knowledge have been proposed, but to date, there is no complete analysis or delineation of
these skills. One contribution of the present study is that it describes a method for eliciting
knowledge about specific complex derivational words. This method makes explicit some word
knowledge that may normally remain implicit, and this provides a measure of MA that is distinct
from the measure of basic morphological skill.
Morphological knowledge is valuable to all readers, but better measures of the skill could be
helpful in identifying and remediating reading problems. There are few published studies on
dyslexia and morphology skill, and the results are not definitive. Several causes of dyslexia have
been put forward, but the prime suspect is a phonological deficit. To some extent, it seems that a
phonological deficit impedes the development of morphology skill (Carlisle, 1987), but it has also
been suggested that dyslexics may use some types of morphemic information to compensate
(Casalis, Cole, & Sopo, 2004). Morphemes are usually larger than phonemes and may be an easier
unit to discern in ongoing speech. On the other hand, morphemes represent a more abstract level
of language than either words or phonemes. In school, there is explicit instruction in letter-sound
pairings and in writing words, but there is much less emphasis on morphemes. It is possible that
dyslexic students derive their knowledge of morphemes from spoken language, and use this level
to compensate for impaired phonemic skills.
Most children master inflectional morphology in the first few years of life, using oral
language. Some derivational morphology is learned from speech, but the most extensive
experience is gained from reading. It would be expected that adult dyslexic readers would have
acquired the basics of word morphology, but that advanced levels of skill would be more elusive
in this group. It would be useful to examine the abilities of adolescent and adult dyslexics to better
understand the clues they utilize in gaining meaning from print. This would require a finely
calibrated test that assesses a clear continuum of morphology skills--another argument for a
morphology test of adult levels of skill.
The US education system has been criticized for its approach to vocabulary, and the
approach seems to be that students are on their own. One education researcher has pointed out that
everything needed for mathematics is taught in math classes, but that when it comes to vocabulary,
students must bring their own word knowledge (Becker, 1977). Children who enter school with
limited vocabularies tend to stay behind in vocabulary skill. This deficit interferes with reading
and can have a detrimental effect on learning in every academic subject. With the enormous
hurdles faced by educators, it is unlikely that vocabulary would be added to the curriculum as an
additional subject, but promoting MA might be feasible. By calling attention to word
relationships, educated adults can guide students in the recognition of morphological relationships.
The fostering of these word analysis skills would give students another tool to use for building
Rarely taught and usually learned implicitly, morphology is beginning to receive more
attention from reading researchers. Reading comprehension requires the fluent, efficient
recognition of words and their meanings, and morphology is surely crucial to this process.
Knowing a word involves more than knowing its definition, and a competent reader must
recognize how a complex word is related to other words. There are countless tests of word
meaning, but there is no standard gauge of word difficulty in terms of morphology, and no test of
this type of knowledge for adult readers. A battery of reading tests typically measures vocabulary,
syntax, and comprehension skills. Morphology is interrelated with all 3 areas, and a single,
relatively short test of MA might allow a quick assessment of the elements that link sounds to
Similarly, accuracy in the self-assessment of vocabulary knowledge would influence one's
approach to reading. Measuring these metalinguistic skills is important for both theoretical and
practical reasons. While the role of PA in learning to read is essentially indisputable, the
contributions of other metalinguistic skills have been proposed. Being able to assess these skills in
adult readers would fill a gap in the research. From a pragmatic point of view, having a quick test
of these skills could help reading teachers pinpoint the particular weaknesses of struggling adult
readers. This study provides one step toward better measurement of these important skills.
APPENDIX: MORPHOLOGY TEST
INSTRUCTIONS: Please check the most appropriate answer for each question.
1. What languages do you speak fluently?
SOther [please list]:
2. In what languages do you read fluently?
SOther [please list]:
Current reading and vocabulary levels
3. During a school semester, how many hours do you spend each week reading for pleasure?
Less than 2 hours
More than 2 hours, but less than 4 hours
More than 4 hours, but less than 6 hours
More than 6 hours, but less than 8 hours
More than 8 hours, but less than 10 hours
More than 10 hours
4. During a school semester, how many hours do you spend each week reading for classes?
Less than 2 hours
More than 2 hours, but less than 4 hours
More than 4 hours, but less than 6 hours
More than 6 hours, but less than 8 hours
More than 8 hours, but less than 10 hours
More than 10 hours
5. In a year, how many books do you read for pleasure?
more than 20
6. Do you think your vocabulary knowledge is adequate for college courses?
SMy vocabulary knowledge is more than adequate for college courses.
SMy vocabulary knowledge is adequate for college courses.
SMy vocabulary knowledge is not adequate for college courses.
7. Do you think your reading ability is adequate for college courses?
SMy reading ability is more than adequate for college courses.
SMy reading ability is adequate for college courses.
SMy reading ability is not adequate for college courses.
Language and reading background
The questions below in #8 refer to difficulties that may have been formally identified through special services or other
testing. If you check Yes for any of the following questions, please explain below.
8. In your native language, have you ever:
had speech therapy?
had a reading problem identified?
had any other language problem identified?
received a diagnosis of a learning difficulty?
been tutored due to difficulty in school?
If yes, in what areas have you had tutoring?
The questions below in #9 refer to problems that you may have noticed, but that were not formally identified.
9. In your native language, have you ever:
had problems with reading?
had problems with spelling?
had any other problems with language?
10. Have you had more than average difficulty learning a foreign language? Yes
11. Does anyone in your immediate family have a learning difficulty? Yes
If yes, who?
Father brothers sisters
If you answered 'yes' to any parts of #8 through 11, please explain:
12. Age: 17
18 19 20
14. Current designation:
Part 1: Look at each word in the list below. Check yes if you know the word. Check no if you do not know the word.
(If you are unsure, check yes if you could use the word in a sentence.)
)o you know the word?
Part 2: In each line, the word in bold font was formed from one of the words on the right. Select the letter of the word
on the right which is the basis for the bold word.
Part 3: Select the best definition for each word.
a. someone in disguise
b. someone not fighting
c. someone with messy hair
a. behaving badly toward
b. clearing of foggy weather
c. unrecognizable food
c. knitted leg covering
a. in the realm of possibility
b. an unlikely story
c. able to survive a tragedy
a. forced to pay with cash
b. a charge that is removed
c. proven false or worthless
a. without a lawsuit
b. with one's bare hands
c. not toned down
a. exchange of letters; matching
b. organisms in a small pond
c. a legal document
a. a 10-year period
b. declining, or self-indulgence
c. varying in pace
a. with a needle
b. using supposition
c. written in ink
a. uneven; not flat
b. outside the solar system
c. giving reasons or causes
a. elf-like, fairy tale character
b. lack of respect
c. baked in a round pan
a. overly bold or confident
b. overly ornate
c. asking over and over
a. place of refuge
b. tutoring in reading
c. entering again
a. in civilian clothes
b. without tiring
c. never losing
a. a lumpy mattress
b. causing trouble or distress
c. intense fear
a. branching out
b. making into song
c. turning prose into poetry
b. going on at the same time
c. changing rapidly
a. causing hatred or dislike
b. cannot be measured
c. a horse out of the barn
a. sharing cooking instructions
b. exchanging or sharing privileges
c. the part left over in division
a. something that can be fixed
b. worthy of praise
c. something that can be sold
a. cannot be undone
b. going forward
c. lacking respect
a. a first job
b. causing a response
c. a mathematical proof
a. carefully chosen
b. food without seasoning
c. viewed with contempt
a. talking too much
a. objecting to something
b. taking a practice test
c. an outdated train or bus depot
a. formal granting of permission
b. making sure of something
c. blaming or criticizing
a. ruling with absolute control
b. removing a stain
c. giving up or surrendering
a. receiving little pay
b. able to afford
c. without equal
b. caring for one's children
a. at the base of
b. an enjoyable puzzle
c. interruption of a pattern
a. setting free
b. sharing books
c. drinking too much
a. digging underground
b. destroying or razing
c. lowering in rank
a. making stronger
b. gaining better eyesight
c. on the spur of the moment
a. shopper or customer
b. someone who takes charge
a. bad stage performance
b. loud behavior
c. editing; preparing for publication
a. to sing in Latin
b. to put into action
c. to perform surgery
a. giving a gift
b. sensing what is about to happen
c. right now; without delay
a. acting with courage
b. cheering loudly
c. shameful behavior
a. unable to choose
a. ancient village
b. moving very slowly
c. not letting light through
a. not stopping
c. without warning
a. lowering of rank
c. without moving
a. unable to be split
b. barely visible
c. in the future
a. making smaller
b. wiping out
a. putting clothes on
b. speaking or writing to
c. outer ornamental layer
a. shaping raw materials
b. smoothing rough edges
c. harmonious interaction
a. finding something new
b. planetary alignment
c. keeping hidden
b. a fixed location
c. a chemical state
a. making well-known
b. speaking badly of
c. hiding from one's fans
LIST OF REFERENCES
Adams, M. J. (1990). Beginning to read: Thinking and learning about print. Cambridge, MA:
Akhtar, N. (2002). Relevance and early word learning. Journal of ChildLanguage, 29, 677-686.
Akhtar, N., Dunham, F., & Dunham, P. J. (1989). Directive interactions and early vocabulary
development: The role of joint attentional focus. Journal of Child Language, 18, 41-49.
Akhtar, N., Jipson, J., & Callanan, M. A. (2001). Learning words through overhearing. Child
Development, 72, 416-430.
Anderson, R. C., & Freebody, P. (1981). Vocabulary knowledge. In J. T. Guthrie (Ed.), Reading
comprehension and education (pp. 77-117). Newark, DE: International Reading
Anderson, R. C., Wilson, P. T., & Fielding, L. G. (1988). Growth in reading and how children
spend their time outside of school. Reading Research Quarterly, 23, 285-303.
Anglin, J. M. (1993). Vocabulary development: A morphological analysis. Monographs of the
Society for Research in Child Development, Serial No. 238, 58(10).
Arnbak, E., & Elbro, C. (2000). The effects of morphological awareness training on the reading
and spelling skills of young dyslexics. Scandinavian Journal of Educational Research, 44,
Baldwin, D. A. (1993). Early referential understanding: Infants' ability to recognize referential
acts for what they are. Developmental Psychology, 29, 832-843.
Baldwin, D. A., & Moses, L. J. (2001). Links between social understanding and early word
learning: Challenges to current accounts. Social Development, 10, 309-329.
Bates, E., Marchman, V., Thal, D., Fenson, L., Dale, P., Reznick, J. S., et al. (1994).
Developmental and stylistic variation in the composition of early vocabulary. Journal of
Child Language, 21, 85-123.
Baumann, J. F., Edwards, E. C., Font, G., Tereshinski, C. A., Kame'enui, E. J., & Olejnki, S.
(2002). Teaching morphemic and contextual analysis to fifth-grade students. Reading
Research Quarterly, 37, 150-176.
Beck, I., & McKeown, M. (1990). Conditions of vocabulary acquisition. In R. Barr, M. L.
Kamil, P. B. Mosenthal, & P. D. Pearson, (Eds.) Handbook ofReading Research, Vol. II
(pp. 789-814). New York, NY: Longman.
Becker, W. C. (1977). Teaching reading and language to the disadvantaged--What we have
learned from field research. Harvard Educational Review, 47, 518-543.
Berko, J. (1958). The child's learning of English morphology. Word, 14, 150-177.
Bianchi, S. M., & Robinson, J. (1997). What did you do today? Children's use of time, family
composition, and the acquisition of social capital. Journal ofMarriage and the Family,
Bloom, P. (2000). How children learn the meanings ofwords. Cambridge, MA: MIT Press.
Bradley, L. & Bryant, P. E. (1983). Categorizing sounds and learning to read: A causal
connection. Nature, 301, 419-421.
Brown, J. I., Fishco, V. V., & Hanna, G. S. (1993). Nelson-Denny Reading Test. Itasca, IL:
Brown, J. I., Fishco, V. V., & Hanna, G. S. (1993). Nelson-Denny Reading Test: Technical
Report Forms G & H. Itasca, IL: Riverside Publishing.
Carlisle, J. F. (1988). Knowledge of derivational morphology and spelling ability in fourth, sixth,
and eighth graders. Applied Psycholinguistics, 9, 247-266.
Carlisle, J. J. (1995). Morphological awareness and early reading achievement. In L. B. Feldman
(Ed.) Morphological aspects of language processing (pp. 189-209). Hillsdale, NJ:
Carlisle, J. F. (2000). Awareness of the structure and meaning of morphologically complex
words: Impact on reading. Reading and Writing: An Interdisciplinary Journal, 12, 169-
Carlisle, J. F. (2003). Morphology matters in learning to read: A commentary. Reading
Psychology, 24, 291-322.
Carlisle, J. F., & Nomanbhoy, D. M. (1993). Phonological and morphological awareness in first
graders. Applied Psycholinguistics, 14, 177-195.
Carroll, J. B., Davies, P., & Richman, B. (1971). Word fequency book. New York: Houghton
Casalis, S., Cole, P., & Sopo, D. (2004). Morphological awareness in developmental dyslexia.
Annals ofDyslexia, 54, 114-138.
Chall, J. S. (1987). Two vocabularies for reading: Recognition and meaning. In M. G.
McKeown & M. E. Curtis (Eds.), The nature of vocabulary acquisition, (pp. 7-17).
Hillsdale, NJ: Erlbaum.
Chall, J. S., & Jacobs, V. A. (2003). Poor children's fourth-grade slump. American Educator, 27,
Chung, T. M., & Nation, P. (2003). Technical vocabulary in specialized texts. Reading in a
Foreign Language, 15, 103-116.
Clark, E. V. (1993). The Lexicon in Acquisition. Cambridge: Cambridge University Press.
Clark, E. V. (1995). Language acquisition: The lexicon and syntax. In J. L. Miller and P. D.
Eimas (Eds.), Speech, Language, and Communication. Handbook of Perception and
Cognition, (2nd ed.) (pp. 303-337). San Diego: Academic Press.
Corson, D. (1997). The learning and use of academic English words. Language Learning. 47,
Coxhead, A. (2000). A new Academic Word List. TESOL Quarterly, 34, 218-238.
Christian, K., Morrison, F. J., Frazier, J. A., & Massetti, G. (2000). Specificity in the nature and
timing of cognitive growth in kindergarten and first grade. Journal of Cognition and
Development, 1, 429-448.
Cunningham, P. M. (1998). The multisyllabic word dilemma: Helping students build meaning,
spell, and read 'big' words. Reading & Writing Quarterly, 141, 189-218.
D'Anna, C. A., Zechmeister, E. B., & Hall, J. W. (1991). Toward a meaningful definition of
vocabulary size. Journal ofReading Behavior, 23, 109-122.
Dale, E. (1965). Vocabulary measurement: Techniques and major findings. Elementary English,
Deacon, S. H., Parrila, R., & Kirby, J. R. (2006). Processing of derived forms in high-functioning
dyslexics. Annals ofDyslexia, 56, 103-128.
Derwing, B. (1976). Morpheme recognition and the learning of rules for derivational
morphology. Canadian Journal of Linguistics, 21, 38-66.
Derwing, B. L., & Baker, W. J. (1979). Recent research on the acquisition of English
morphology. In P. Fletcher and M. Garman (Eds.) Language acquisition: Studies infirst
language development (pp. 209-223). Cambridge, UK: Cambridge University Press.
Dolch, E. (1936). How much word knowledge do children bring to grade one? Elementary
English Review, 13, 177-183.
Donahue, P. L., Finnegan, R. J., Lutkus, A. D, Allen, N. L., & Campbell, J. R. (2001). The
nation's reading report card: Fourth-grade reading 2000. National Assessment of
Educational Progress: Washington, D. C.
Duke, N. K. (2000). 3.6 minutes per day: The scarcity of informational texts in first grade.
Reading Research Quarterly, 35, 202-224.
Fowler, A. E., & Liberman, I. Y. (1995). The role of phonology and orthography in
morphological awareness. In L. B. Feldman (Ed.) Morphological aspects of language
processing (pp. 157-188). Hillsdale, NJ: Erlbaum.
Freyd, P., & Baron, J. (1982). Individual differences in acquisition of derivational morphology.
Journal of Verbal Learning and Verbal Behavior, 21, 282-295.
Gaustad, M. G., & Kelly, R. R. (2004). The relationship between reading achievement and
morphological word analysis in deaf and hearing students matched for reading level.
Journal ofDeaf Studies and DeafEducation, 9, 269-285.
Gleitman, L. R. (1994). Words, words, words... Philosophical Transactions: Biological
Sciences, 346, 71-77.
Golinkoff, R. M., Shuff-Bailey, M., Olguin, R. (1995). Young children extend novel words at the
basic level: Evidence for the principle of categorical scope. Developmental Psychology,
Goodman, J. C., McDonough, L., Brown, N.B. (1998). The role of semantic context and memory
in the acquisition of novel nouns. ChildDevelopment, 69, 1330-1344.
Goulden, R., Nation, P., & Read, J. (1990). How large can a receptive vocabulary be? Applied
Linguistics, 11, 341-363.
Hall, D. G., & Graham, S. A.(1999). Lexical form class information guides word-to-object
mapping in preschoolers. ChildDevelopment, 70, 78-91.
Hayes, D. P. (1988). Speaking and writing: Distinct patterns of word choice. Journal of Memory
and Language, 27, 572-585.
Hayes, D., Wolfer, L. T., & Wolfe, M. F. (1996). Schoolbook simplification and its relation to
the decline in SAT-verbal scores. American Educational Research Journal, 33, 498-508.
Hirsh-Pasek, K., Golinkoff, R. M., & Hollich, G. (2000). An emergentist coalition model for
word learning: Mapping words to objects is a product of the interaction of multiple cues.
In R. M. Golinkoff, K. Hirsh-Pasek, L. Bloom, L. B. Smith, A. L. Woodward, N. Akhtar,
M. Tomasello, and G. Hollich (Eds.), Becoming a word learner: A debate on lexical
acquisition (pp. 136-164). Oxford, England: Oxford University Press.
Hoff, E., & Naigles, L. (2002). How children use input to acquire a lexicon. Child Development,
Hofferth, S. L., & Sandberg, J. F. (2001). How American children spend their time. Journal of
Marriage and Family, 63, 295-308.
Hurry, J., Nunes, T., Bryant, P., Pretzlik, U., Parker, M., Curno, T.., & Midgley, L. (2005).
Transforming research on morphology into teacher practice. Research Papers in
Education, 20, 187- 206.
Juel, C. (1988). Learning to read and write: A longitudinal study of 54 children from first
through fourth grades. Journal of Educational Psychology, 80, 437-447.
Katusic, S. K., Colligan, R. C., Barbaresi, W. J., Schaid, D. J. (2001). Incidence of reading
disability in a population-based cohort, 1976-1982, Rochester, Minn. Mayo Clinic
Proceedings, 76, 1081-1092.
Kaye, D. B., Sternberg, R. J., & Fonseca, L. (1987). Verbal comprehension: The lexical
decomposition strategy to define unfamiliar words. Intelligence, 11, 1-20.
Kelly, R. R., & Gaustad, M. G. (2007). Deaf college students' mathematical skills relative to
morphological knowledge, reading level, and language proficiency. Journal ofDeaf
Studies and DeafEducation, 12, 25-37.
Kemp, N. (2006). Children's spelling of base, inflected, and derived words: Links with
morphological awareness. Reading and Writing, 19, 737-765.
Kidd, E., Lieven, E., & Tomasello, M. (2006). Examining the role of lexical frequency in the
acquisition and processing of sentential complements. Cognitive Development, 21, 93-
Kuo, L. & Anderson, R. C. (2006). Morphological awareness and learning to read: A cross-
language perspective. Educational Psychologist, 41, 161-180.
Libben, G., & Jarema, G. (2004). Conceptions and questions concerning morphological
processing. Brain and Language, 90, 2-8.
Lorge, I., & Chall, J. (1963). Estimating the size of vocabularies of children and adults: An
analysis of methodological issues. The Journal ofExperimental Education, 32, 147-157.
Mahony, D., Singson, M., & Mann, V. (2000). Reading ability and sensitivity to morphological
relations. Reading and Writing: An Interdisciplinary Journal, 12, 191-218.
Markman, E. M. (1990). Constraints children place on word meanings. Cognitive Science, 14,
Markson, L., & Bloom, P. (2001). Evidence against a dedicated system for word learning in
children. In M. Tomasello and E. Bates (Eds.), Language development: The essential
readings (pp. 130- 133). Malden, MA: Blackwell.
Marslen-Wilson W., Tyler, L. K., Waksler, R., & Older, L. (1994): Morphology and meaning in
the English mental lexicon. Psychological Review, 101, 3-33.
Masterson, J., & Hayes, M. (2004). UK data from 197 undergraduates for the Nelson-Denny
Reading Test. Journal ofResearch in Re, ,ding. 27, 30-35.
Maylath, B. (1997) Why do they get it when I say 'gingivitis' but not when I say 'gum
swelling'? New Directionsfor Teaching and Learninug. 70, 29-37.
McKinnon, R., Allen, M., & Osterhout, L. (2003). Morphological decomposition involving non-
productive morphemes: ERP evidence. Cognitive Neuroscience andNeuropsychology,
McBride-Chang, C., Wagner, R. K., Muse, A., Chow, B. W.-Y., & Shu, H. (2005). The role of
morphological awareness in children's vocabulary acquisition in English. Applied
Psycholinguistics, 26, 415-435.
Merriam-Webster (2006-2007). New words. Retrieved on February 1, 2007, from Merriam-
Webster Online from http://www.m-w.com/info/new words.htm
Miller, G. M., & Wakefield, P. C. (1993). On Anglin's analysis of vocabulary growth analysis.
Monographs of the Society for Research in Child Development, Serial No. 238, 58, 167-
Moats, L. C., & Smith, C. (1992). Derivational morphology: Why it should be included in
language assessment and instruction. Language, Speech, and Hearing Services in
Schools, 23, 312-319.
Moore, C., Angelopoulos, M., & Bennett, P. (1999). Word learning in the context of referential
and salience cues. Developmental Psychology, 35, 60-68.
Nagy, W. E., & Anderson, R. C. (1984). How many words are there in printed school English?
Reading Research Quarterly, 19, 304-330.
Nagy, W., Anderson, R. C., Schommer, M., Scott, J. A., & Stallman, A. C. (1989).
Morphological families in the internal lexicon. Reading Research Quarterly, 26, 262-
Nagy, W. E., & Anderson, R. C. (1995). Metalinguistic awareness and literacy acquisition in
different languages. Technical Report No. 618, Center for the Study of Reading,
University of Illinois at Urbana-Champaign.
Nagy, W., Berninger, V. W., & Abbot, R. D. (2006). Contributions of morphology beyond
phonology to literacy outcomes in upper-elementary and middle-school students. Journal
ofEducational Psychology, 98, 134-147.
Nagy, W., Berninger, V., Abbott, R., Vaughan, K., & Vermeulen, K. (2003). Relationship of
morphology and other language skills to literacy skills in at-risk second-grade readers and
at-risk fourth-grade writers. Journal ofEducational Psychology, 95, 730-742.
Nagy, W. E., Diakidoy, I-A. N., & Anderson, R. C. (1993). The acquisition of morphology:
Learning the contribution of suffixes to the meanings of derivatives. Journal of Reading
Behavior, 25, 155-170.
Nagy, W. E., & Scott, J. A. (2000). Vocabulary processes. In M. L. Kamil, P. B. Mosenthal, P.
D. Pearson, & R. Barr (Eds.), Handbook ofReading Research, Volume III (pp. 269 -284).
Mahwah, NJ: Erlbaum.
Nation, I. S. P. (2001). Learning vocabulary in another language. Cambridge: Cambridge
Nelson, K. (1988). Constraints on word learning. Cognitive Development, 3, 221-246.
Nunes, T., Bryant, P., & Bindman, M. (1997). Morphological spelling strategies: Developmental
stages and processes. Developmental Psychology, 33, 637-649.
Nunes, T., Bryant, P., & Bindman, M. (2006). The effects of learning to spell on children's
awareness of morphology. Reading and Writing, 19, 767-787.
Olguin, R. & Tomasello, M. (1993). Twenty-five-month-old children do not have a grammatical
category of verb. Cognitive Development, 8, 245-272.
Ouellette, G. P. (2006). What's meaning got to do with it? The role of vocabulary in word
reading and reading comprehension. Journal of Educational Psychology, 98, 544-566.
Pinker, S. (1994). The Language Instinct. New York: Harper Collins.
Samuelson, L. K., & Smith, L. B. (1998). Memory and attention make smart word learning: An
alternative account of Akhtar, Carpenter and Tomasello. Child Development, 69, 94-104.
Savage, R., Blair, R., & Rvachew, S. (2006). Rimes are not necessarily favored by prereaders:
Evidence from meta- and epilinguistic phonological tasks. Journal of Experimental Child
Psychology, 94, 183-205.
Shaywitz, S. E., & Shaywitz, B. A. (2005). Dyslexia (Specific reading disability). Biological
Psychiatry, 57, 1301-1309.
Shibles, B. H. (1959). How many words does a first grade child know? Elementary English, 36,
Singson, M., Mahony, D., & Mann, V. (2000). The relation between reading ability and
morphological skills: Evidence from derivational suffixes. Reading and Writing: An
interdisciplinary journal, 12, 219-252.
Smith, L. B. (1999). Children's noun learning: How general learning processes make specialized
learning mechanisms. In B. MacWhinney (Ed.), The emergence of language, (pp. 277-
303). Mahwah, NJ: Erlbaum.
Stanovich, K. E. (1986). "Matthew effects" in reading: Some consequences of individual
differences in the acquisition of literacy. Reading Research Quarterly, 4, 360-407.
Tomasello, M. (2001). The item-based nature of children's early syntactic development. In M.
Tomasello & E. Bates (Eds.) Language development: The essential readings (pp. 169-
186). Malden, MA: Blackwell.
Tomasello, M. (1995). Joint attention as social cognition. In C. Moore & P. J. Dunham (Eds.),
Joint attention: Its origins and role in development, (p.103-130). Hillsdale, NJ:
Tomasello, M., & Barton, M. (1994). Learning words in nonostensive contexts. Developmental
Psychology, 30, 639-650.
Tomasello, M., & Farrar, M. J. (1986). Joint attention and early language. ChildDevelopment,
Topping, K. J., & Paul, T. D. (1999). Computer-assisted assessment of practice at reading: A
large scale survey using Accelerated Reader data. Reading & Writing Quarterly, 15, 213-
Tyler, A., & Nagy, W. (1989). The acquisition of English derivational morphology. Journal of
Memory and Language, 28, 649-667.
Verhoeven, L., & Carlisle, J. F. (2006). Introduction to the special issue: Morphology in word
identification and word spelling. Reading and Writing, 19, 643-650.
Wagner, K. R. (1985). How much do children say in a day? Journal of ChildLanguage, 12,
Wang, M., Cheng, C., & Chen, S.-W. (2006). Contribution of morphological awareness to
Chinese-English biliteracy acquisition. Journal of Educational Psychology, 98, 542-
White, T., Power, M., & White, S. (1989). Morphological analysis: Implications for teaching
and understanding vocabulary growth. Reading Research Quarterly, 24_283-304.
Windsor, J., & Hwang, M. (1999). Derivational suffix productivity for students with and without
language-learning disabilities. Journal of Speech, Language, and Hearing Research, 42,
Wysocki, K., & Jenkins, J. R. (1987). Deriving word meanings through morphological
generalization. Reading Research Quarterly, _22, 66-81.
Lisa Kay Maag was born in Pensacola, Florida, and she grew up in the Washington, D. C.
area and in Southern California. Her original career goal was to be a botanist, and she earned a
Bachelor of Science degree in biology, and later, a Master of Business Administration degree.
She worked mainly in small business management, but also spent a few years as a high school
science teacher. Her father had earned a Ph.D. in psychology at the University of Florida, and
she initially had no intention of studying psychology, but eventually she returned to Florida and
decided to pursue her interest in language development and its relation to cognitive skills. This
led to graduate school at the University of Florida, and (after an interruption of several years), to
earning a Ph.D. in developmental psychology. She is a faculty member of the University of
Phoenix, where she teaches psychology and related courses. She plans to continue her research in
the development of morphological awareness and vocabulary and to continue to collect a wide
variety of flowering plants.