What's in a picture? Comparing gender constructs of younger and older adults.
Abstract: To identify tacit gender role constructs, 192 younger adults (M age = 21.1) and 126 older adults (M age = 70.1) were given a series of six drawings unrelated to gender, asked to create a brief description of the person who made each drawing, and then asked to guess the person's age and sex. Using the Consensual Qualitative Research (CQR) system as a guide, gender prototypes were developed. Three findings were significant. First, there was widespread agreement across generations and across sexes on the central features of prototypical femininity (low self-esteem, restricted/ oppressed, emotional distress) and on prototypical masculinity (successful/accomplished, risk-taking, strong/determined, problem-solving, and emotional distress). Second, significant intergenerational differences regarding masculinity were found (younger adults citing more role strain and more emotional expressiveness). Third, one within-generation difference was found (younger men versus younger women regarding prototypical femininity). Implications of these findings are discussed.

Key Words: masculine traits, feminine traits, gender stereotypes, gender role constructs, gender role transition
Subject: Adults (Social aspects)
Adults (Psychological aspects)
Femininity (Social aspects)
Masculinity (Social aspects)
Sex role (Social aspects)
Youth (Social aspects)
Youth (Psychological aspects)
Authors: Robertson, John M.
Johnson, Ann L.
Benton, Stephen L.
Janey, Bradley A.
Cabral, Jennifer
Woodford, Joyce A.
Pub Date: 09/22/2002
Publication: Name: The Journal of Men's Studies Publisher: Men's Studies Press Audience: Academic Format: Magazine/Journal Subject: Social sciences; Women's issues/gender studies Copyright: COPYRIGHT 2002 Men's Studies Press ISSN: 1060-8265
Issue: Date: Fall, 2002 Source Volume: 11 Source Issue: 1
Topic: Event Code: 290 Public affairs
Product: Product Code: E121940 Adults; E121930 Youth
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 94077706
Full Text: Researchers long have been intrigued by popular assertions that women are more "--," or that men are more "--." Over the last 30 years, the challenge Of identifying reliable constructs about gender has stimulated both theoretical speculation (e.g., Chesler, 1972; David & Brannon, 1976; Goldberg, 1976; Pleck, 1981), and a continuing flow of published research.

Two types of questions have been central to these endeavors. The first set of questions has addressed measurement issues. Which research designs are more likely to discover any constructs that might distinguish femininity from masculinity? Or, are these variables so overlapping and continuous as to make any distinctions unreliable? What methodology is most likely to answer these questions?

A second cluster of questions has focused more on the content of hypothesized differences. Are gender construct differences primarily about personality traits, individual interests, physiological processes, or other attributes? Is gender best understood as a single dimension with masculinity and femininity located at opposite ends of the same continuum? Or, are femininity and masculinity better understood as two independent variables?


Helgeson's (1994) brief review of the history of this issue noted that one of the first modern attempts to measure differences between females and males was made by Terman (best known for his work in defining the Intelligence Quotient or IQ) and one of his graduate students. Together, Terman and Miles (1936) developed an instrument they called the Attitude Interest Analysis Survey, which asked subjects to make word associations, examine inkblots, report emotional responsiveness, and express their attitudes about various occupations, books, and famous people. Item scores on this instrument were examined to discover any gender differences, and these items were then defined as either feminine or masculine depending on which sex scored higher. During the next few years, other well-known instruments published sex differences in scores; note, for example, Strong's (1936) Vocational Interest Blank and the Minnesota Multiphasic Personality Inventory (Hathaway & McKinley, 1940). For all these instruments, the gender-related scale consisted of items that discriminated male scores from female scores. An underlying theory about gender simply was not central to the development or purpose of these instruments.

Most of these early measures of gender differences suffered from several limitations: they did not assess masculinity and femininity as theoretically defined concepts; they did not speak to the possibility that gender similarities may be greater than gender differences; they obfuscated the difference between gender and sexual orientation; and they represented femininity and masculinity as opposite ends of a single dimension called gender.

These limitations were addressed in the 1970s by the development of two new measurement instruments: the Bern Sex-Role Inventory (BSRI) and the Personal Attributes Questionnaire (PAQ). Bem (1974) challenged the idea that masculinity (M) and femininity (F) were simply opposite ends of a single dimension and argued that a person could be high or low on both M and F. Masculine items were defined as characteristics more desirable in males than females, and feminine traits as more desirable in females than males. This conceptual shift was significant; no longer would researchers use only a frequency count of miscellaneous items to define masculine and feminine differences. Rather, the definition moved to items that were identified as desirable traits for males and females.

Similarly, Spence and her colleagues (Spence & Helmreich, 1978; Spence, Helmreich, & Stapp, 1974) based the PAQ on the idea that femininity and masculinity were independent constructs. Again, items were assigned to F or M scales based partly on their desirability as personality traits, and partly on how characteristic each trait was of men or women. The PAQ also presented an M-F scale, which included socially desirable items on which the ratings of ideal males and females were different. A subsequent version (EPAQ) included both positive and negative descriptors of femininity and masculinity (Spence, Helmreich, & Holahan, 1979). Both the BEM and the PAQ selected personality characteristics for the items in their instruments. Subsequently, these dimensions were defined more as "instrumentality" and "expressiveness," rather than masculinity and femininity (Spence, 1984).

More recently, other contributions have been made to the discussion about the measurement of gender constructs. For example, definitions of femininity and masculinity were expanded to include not only personality traits, but also physical appearance and actual behavior (e.g., Myers & Gonda, 1982). Twenge (1999) expanded the number of domains even further and recommended seven possible factors in thinking about gender: instrumental, occupational, physical, stylistic, relational, attitudinal, and self-ascribed elements.

Other recent attempts to measure gender constructs have used open-ended methodologies, such as self-report strategies (e.g., Helgeson, 1994; Martin, 1987; Pennell & Ogilvie, 1995), behavioral observations (e.g., Allen, 1995), meta-analyses (e.g., Swim, 1994), and cross-cultural comparisons of tasks (e.g., Condon & Stern, 1993). Consistent with the "bubble hypothesis" that for every methodological solution, there is a cost to be paid elsewhere (Gelso, 1979), these approaches have had several obvious drawbacks. Self-reports are influenced by social desirability and impression management (Judd & Park, 1993). Behavioral observations can be colored by biases of the observers, by the laboratory setting in which the observations are made, or by the demonstrated tendency to view self and others differently (the actor/observer effect; see Jones & Nisbett, 1972). Meta-analyses are limited to an examination of traits actually included in the studies chosen for the meta-analysis; and cross-cultural studies tend to focus more on tasks and activities (such as crafts and manufacturing, household duties, initiation rites, and warfare) than on personality traits (cf., Ember & Ember, 1971; Munroe, Munroe, & Whiting, 1981; Murdock & Provost, 1973).

Three design strategies avoid some of these limitations: the personal constructs approach, the prototype design, and the use of drawings. These approaches aim more at eliciting constructs directly from subjects rather than providing descriptive items to subjects for evaluation. First, the personal construct approach is illustrated by Baldwin, Critelli, Stevens, and Russel (1986), who adapted their ideas from Kelly's (1955) Role Construct Repertory Test. Subjects were asked to make comparisons among several persons: the most feminine person they knew, the most masculine person they knew, themselves, and nine other people. The instructions asked people to describe how two people were alike, but different from a third. The advantage of this approach was that it enabled people to reveal their own constructs about gender without limiting their responses to an evaluation of typical dichotomies or adjective lists. The result was a more individualized interpretation of sex role differences.

Second, based on Locksley and Colten's (1979) work, the prototype strategy assumes that masculinity and femininity are not sharply defined categories at all. Rather, they are ideas that have indistinct boundaries organized around a loosely defined model. The prototype can be either a perfect example of the category or a description that contains central descriptive features. Prototypes are based on the idea that people use many different elements in defining notions such as gender. Some of these factors are close to the center of the prototype, meaning that many different people are likely to use that same factor. Other descriptors are more idiosyncratic, and therefore likely to be used more infrequently. Using this approach, Helgeson (1994) reported a prototype that included about 20 of the most common features that her subjects ascribed to femininity and masculinity. The notion of prototype has also been used more loosely in exploring masculinity traits in popular literature (Highet, 1998), in portraying femininity in the context of psychosomatic distress (Luca-Stolkin, 1999), and in examining the role of face recognizability in gender prototypes (O'Toole, Defenbacher, Valentin, McKee, Huff, & Abdi, 1998).

Third, an even more indirect method to reveal gender constructs is the use of drawings. A variety of techniques have been used to explore gender constructs in this way. Condon and Stem (1993) developed a "draw-a-person-interview" procedure, in which gender-related information was elicited by asking subjects to explain drawings they would make of a man or woman. Other studies have asked subjects to draw a person (based on the work of Machover, 1949) and then have analyzed such variables as the sex of the drawn figure, interactions between the sex of the subject and the experimenter, and scores on selected aspects of gender (e.g., Houston & Terwilliger, 1995). Researchers have also used drawings of research participants to analyze measurable differences in sketches of body parts (Ortega, Arnold, & Smeltzer, 1988), to compare female and male depression scores (Holmes & Wiederholt, 1982), to elicit children's constructs about the frequency and intensity of emotional expression in males and females (Karbon, Fabes, Carlo, & Martin, 1992), and to reveal stereotypes in the interpretation of animal behavior in children's books (White, Morris, & Arthur, 1996). Drawings have also been used in cross-cultural studies to compare gender constructs of children in Boston and Israel (Rubenstein, Feldman, Rubin, & Noveck, 1987), and to compare gender ratings of kibbutz-born females and males in Israel (Snarey & Son, 1986).

The use of drawings offers several advantages in exploring gender constructs. Drawings allow people to offer their own conceptions of gender, and therefore avoid the limitations of checklists or prior theoretical considerations. It is therefore more likely that the constructs people actually use in thinking about gender will emerge.

All of these studies using drawings have had one element in common. They have asked direct and explicit questions about gender. Examples include, "Is this drawing a woman or a man?" "Plebe draw two boys doing something together." Or, "How would you describe a feminine female?" The risks associated with social desirability and impression management come into play when it is clear to subjects that the question of interest to the researcher is about gender.

Although the use of drawings to explore traits of humans has been heavily criticized when used for certain purposes (Lilienfeld, Wood, & Garb, 2000), mostly because of variable reliability and low validity, these criticisms are directed at the proposition that trained judges can make accurate observations about people based only on their drawings. Another use of drawings can be simply to elicit the underlying constructs of the person commenting on the drawing; this approach is built more on personal construct theory (Kelly, 1955) than on the theories of art therapy.

Given these considerations, a promising approach to the assessment of differences in constructs about femininity and masculinity would combine these elements: it would be open-ended and allow any available construct to emerge (no checklists or grids); and it would be indirect in that subjects would not be actively cued that the research was related to gender. We have found only one study that used a similar approach. Raty and Snellman (1997) asked children to draw an "ordinary" person and an "intelligent" person, and found that most children drew an "intelligent" person as an adult male.


Many attempts to distinguish masculine constructs from feminine constructs have concluded that men tend to be more focused on impersonal tasks and activities, while women are more interested in social interactions and relationships. The language used to define this distinction has included instrumentality and expressivity (Parsons & Bales, 1955), agency and communion (Bakan, 1966), justice and care (Gilligan, 1982), and autonomy and connectedness (Markus & Oyserman, 1989).

These contrasting constructs have been developed into checklists. For example, the BSRI (Bern, 1974) description of masculinity included such features as leadership and aggression. The PAQ (Spence & Helmreich, 1978; Spence, Hehmreich, & Holahan, 1979) listed both positive traits (e.g., independence and decisiveness), and negative traits (e.g., cynical and greedy). Similar lists of features were developed for femininity. The BSRI included items like gentle and understanding. The PAQ listed positive features (e.g., devoted to others and kindness), as well as negative traits (e.g., spineless and nagging). Evidence suggests that these lists are reliable, in that subsequent studies have found that they do distinguish females and males from each other (e.g., Martin, 1987).

The BEM and the PAQ have one thing in common: All the items can be considered personality variables. It is not surprising that more open-ended design strategies have created much broader descriptions of the two genders. The most prominent discovery was that people do not limit their understanding of gender to personality traits, as these lists imply. Two studies illustrated what happens when people are given the opportunity to generate their own descriptions. Helgeson (1994) asked subjects to describe one of six stimulus persons (masculine person, masculine male, masculine female; and feminine person, feminine male, or feminine female). In effect, her work developed a prototype of the constructs of the two genders, using ideas generated by her subjects. She found that femininity included not only personality traits (such as caring, delicate, and shy), but also interests (a concern with appearance, likes art), and physical appearance features (well dressed, attractive, smiles). Masculinity likewise included personality traits (self-confident, dominant), interests (sports, cars), and physical descriptions (muscular, tall, hairy face). What Helgeson demonstrated was that people include many elements in their understanding of gender beyond personality traits. She also found that people had many idiosyncratic ideas about gender; that is, many descriptors appeared in her sample only rarely. This finding lent support to the notion that using prototypes is a helpful way to approach an understanding of gender constructs, and that the development of gender-specific checklists may have more limits than benefits.

A second illustration of what happens with an open-ended approach is a study that used Jack-O-Lantern drawings to elicit gender-related notions (Lott, 1979). Drawings of Jack-O-Lanterns by kindergartners were given to adults who were asked to identify the sex of the artist and provide reasons for the decision. Judges decided that a girl made the drawing when it was neat, colorful, smiling, and symmetrical; and that a boy made the drawing when it was messy, frightening, incomplete, and unconventional. Clearly, these elements are part of the view adults have of gender, yet they are not included in the standard instruments that measure femininity and masculinity.

Another important variable in thinking about gender is the role of age. Given the common perception that we are living through a time of gender role transitions, it might be hypothesized that different age groups might have different views of femininity and masculinity. It is not surprising, therefore, to find studies using drawings to explore age as a variable in understanding gender constructs. Silver (1987) found differences in the emotional content of drawings that spanned four age groupings--third graders, seniors in high school, young adults, and older adults. Results across all age groups suggested that males more frequently saw themselves as having to fight in dangerous environments, and females saw themselves as part of their environment rather than living in opposition to it.

The study of gender differences has not been without its critics. Given the diversity of traits that have been found in the literature, some have argued that it may no longer be socially relevant to use the categories of masculine and feminine (Constantinople, 1973; Gergen, 1991). Others have noted that the gender similarities outweigh the differences (Tavris & Wade, 1984). Still others have found that cultural sex stereotyping is more extreme than actual group norms warrant (Martin, 1987), and that people think differently about gender differences, depending on whether they are describing ideal women and men, their own views of the differences, or society's views (Hort, Fagot, & Leinback, 1990; Pennell & Ogilvie, 1995). Nevertheless, as noted above, basic distinctions continue to be discussed (e.g., Philpot, 2000).

A practical matter makes the continued study of gender constructs potentially useful. A number of significant health issues appear to be influenced by gender related behavior. To illustrate, females experience more overt depression, loneliness (Wheeler, Reis, & Nezlek, 1983), and distress (Frank, McLaughlin, & Crusco, 1984) than men. At the same time, Courtenay (2000) cited U.S. Department of Health and Human Services figures, which show that men die nearly seven years earlier than women and have death rates that are higher for all 15 leading causes of death. These differences may be related to lifestyle differences, not biological differences. A recent review (cited in Courtenay, 2000) indicated "men are more likely than women to engage in more than 30 behaviors that increase the risk of disease, injury, and death" (p. 5). These rather pragmatic considerations suggest that continuing to explore gender constructs may lead to some beneficial applications.

In sum, the literature exploring the content of gender constructs suggests that: femininity and masculinity appear to be independent dimensions, not opposite ends of a single dimension; gender constructs include beliefs about personality traits, interests, physical appearance, and perhaps other factors; and the use of prototype and drawing methodologies provide a view that is closer to people's experience of gender than do checklists and theories. Underscoring the importance of continuing to study these issues is the apparent connection between gender and mental/physical health and the continuing awareness that we are in a period of gender role transition.


This study had two purposes. The first was to elicit information about the constructs of gender, but to do so obliquely; that is, we wanted to gather information about people's views of gender without asking them directly. The second purpose was to determine whether this approach would uncover any differences in the ways younger adults and older adults think about gender.



Two groups of participants were recruited, based on age. Young adults were students recruited from classes in the departments of education and psychology at a Midwest university campus. Older adults were recruited from several sources, including a retired teachers association, a retired federal employees group, a senior advocacy organization, church groups for older adults, professional women's organizations, and members of ambulatory retirement communities.

The younger adult sample had 192 participants (121 females, 71 males). The mean age for the younger females was 21.0 (SD = 2.7), and for the younger males was 21.4 (SD = 2.9). The older adult sample had 126 participants (68 females, 58 males). The mean age for the older females was 72.6 (SD = 10.3), and for the older males was 69.5 (SD = 12.1).

Overall, the participants were mostly European American (90%). There was no significant ethnic difference among the four groups. Expressed in rounded percentages, ethnic minorities identified themselves as African Americans (1%), Hispanic Americans (1%), Asian Americans (1%), Arabic Americans (.5%), Native Americans (5%), and "Other" (2%).

We asked about relationship status. Among the younger males, 71.8% were single, 11.3% were partnered in a relationship, 11.3% were married, and 4.2% were engaged. Among the younger females, 53.7% were single, 21.5% were partnered, 14.9% were married, and 9.9% were engaged. Among the older males, 74.1% were married, 19.0% were widowed, and less than 2% each were single, engaged, divorced, or partnered. And, among the older females, 50.0% were married, 38.2% were widowed, 5.9% were single, 4.4% were divorced, and less than 2% were partnered.


Drawings. A booklet with six drawings was presented to each participant (Figure 1; see Appendix A). The drawings were not overtly related to gender, but were described as illustrations of "How People Cope with Ongoing Challenges." In point of fact, the explanation accurately described how the drawings were created. A group of university students with learning disabilities had been identified through a student services office, and had been invited by letter to participate in a study exploring various aspects of their experiences. Part of the study had invited participants to draw a picture that illustrated how they had coped with their challenges (cf., Johnson, 1997). No information about gender was included in the instructions.

Six drawings from this collection were selected for the present study. Selection criteria included intelligibility of the drawings and their captions, generalizability beyond learning disability issues, sex balance (using three drawings by women, three by men), and a diversity of topics addressed in the drawings. Using these criteria, six drawings were selected by a consensus of four full-time doctoral level clinicians who consulted with a focus group of undergraduate students. Each drawing included a title and a brief explanatory caption provided by the person who made the drawing.

Participants in the present study were asked to examine each drawing and then answer the question, "What words would you use to describe the person who made this drawing? You may consider anything that comes to mind--personality traits, interests, physical appearance, typical behavior--whatever seems relevant to you."

Two pilot studies of the drawings were conducted, and the qualitative results were compared with the consensus criteria. These studies led to a decision to drop one drawing and to replace it with another to increase the breadth and clarity of the issues raised by the drawings.

Age and sex questions. Two questions were included for each drawing, and were asked after subjects had completed their written descriptions of the persons who made the drawings. We asked participants to guess the age and sex of the people who made the drawings, and to state why. With regard to the age question, younger adults tended to guess lower ages (M = 20.7, SD = 9) than older adults (M = 30.9, SD = 13.4). When guessing the sex of the person who made the drawings, both groups tended to select male more often than female (younger adults = 63%; older adults = 62%). Broken down by drawing, younger and older adults agreed that males made three of the drawings: Drawings 2 (83%), 5 (68%), and 6 (71%). Older and younger participants split down the middle on Drawing 3. Younger adults thought Drawing 1 was made by a male (54%), but older adults guessed female (55%). And on Drawing 4, younger adults guessed female (53%), while older adults were evenly split.

To determine if our interest in gender constructs was detected, we added a final question asking participants to report what they believed might be learned from the study. In the entire sample of 318 research participants, 15 (4.7%) correctly guessed that we were interested in understanding more about gender constructs. Most participants reported that they had no idea what we were studying; among those who did guess, the most common suggestion was that we were interested in understanding more about human problems or how people deal with challenges. Without knowing our interest in gender ideas, participants were not subject to social desirability considerations, and so we concluded that we were likely tapping tacit gender role constructs.


Data were collected in groups in several settings. The young adult data were collected separately from the older adult data. Participants were gathered in groups and given the 14-page questionnaires. A standardized set of instructions for completing the questionnaires was given to them. Completion time varied, but generally averaged about 45 minutes.

To analyze the qualitative data, we adapted procedures from the Consensual Qualitative Research (CQR) system developed by Hill and her collaborators (Hill, Thompson, & Williams, 1997). The CQR approach has been used to examine several topics related to the process of psychotherapy (e.g., Hill, Nutt-Williams, Heaton, Thompson, & Rhodes, 1996; Knox, Hess, Petersen, & Hill, 1997; Ladany, O'Brien, Hill, Melincoff, Knox, & Petersen, 1997). Because of its success in classifying large amounts of qualitative data into meaningful categories, the CQR was adapted for use in this study.

The CQR system recommends three steps for conducting an analysis of data collected from open-ended questions: developing domains and coding them, constructing core ideas from the domains, and developing categories that show consistencies across the cases in the sample. At each stage in the process, the central goal is to develop a consensus among teams of raters.

We organized our research team into groups of three or four persons each to conduct the analyses. The advantage of using several persons to evaluate the data was that the bias of one person did not become unduly influential. Reaching consensus within a team and using auditors to look at the conclusions of each group required that group members listen carefully to each other and share their views with a sense of mutual respect and power.

CQR is not concerned with calculating inter-rater reliability. Rather, the process is driven by the need to be careful and reasoned with regard to the opinions each expresses, and to be adaptive when it seems appropriate. It should be noted that this approach can create intense and energetic discussions, as people present their views and listen to others with contrasting views. Finding a consensus is sometimes easy, and sometimes very challenging. Styles (1993) has argued that "qualitative research shifts the goal of quality control from the objective truth of statements to understanding by people" (p. 593). Consistent with his view, the primary task of our team members was to reach a shared understanding of the data and to organize it in meaningful ways.

In our study, the raw data to be understood were the collection of thousands of words, phrases, and sentences used by participants when answering the question, "What words would you use to describe the person who made this drawing?" Following the CQR approach, our first step was to develop meaningful domains of ideas from these data and to code these domains. We began by separating all the handwritten answers to the stimulus question into categories defined by sex and age. The views of younger and older research participants were thus separated, as well as the views of female and male participants. These lists contained hundreds of descriptors that needed to be examined for common themes.

Each team member began by working alone, examining the data, and looking for similarities in meaning among the various words and phrases written on the questionnaires. The objective was to find all the descriptors that seemed to identify the same general trait or characteristic. In this way, domains were formed that contained a list of all the words our research participants used when describing that feature. When participants wrote longer sentences or paragraphs, our team member would suggest a summary word or phrase.

The second step was for each person to present his or her suggestions of domains to the team. The team then reviewed all the suggestions and worked to create a consensus on the label and the content for each domain. For example, a team might agree to create the domain "Strong" from such descriptors as "strong," "strong-willed," "staying strong," "does not seem weak," and "wanting to remain strong." Another domain might be called "Planful," and would include such descriptors as "looks down the road," "wants to organize his future," "likes to have things prepared," and so forth. Team members considered various suggestions for revising or refining the domains. Recommendations included moving a descriptor from one domain to another, re-naming a domain, separating a domain into two or more new domains, or merging various domains. The purpose of this process was to reach agreement on naming the domains and to place all of the descriptors into an appropriate domain. This procedure continued until group members had no more suggestions. The result was an inventory of all the ideas that emerged as our younger and older adults described the persons whom they believed made the drawings.

Auditors from another group then reviewed the domains that had been developed. The auditors suggested revisions, and the original team members edited and re-organized their lists until a consensus was reached. When this process resulted in an agreement, the list was regarded as complete.

The third step was to develop names for the domains that were consistent across various segments of our sample. To illustrate, the team examining the descriptors given by younger males might have named a domain as "Adventuresome," whereas another team examining a similar list given by older males might have named the same features as "Risk-taking." In order to facilitate the task of comparing the views of older and younger adults, we wanted to use the same label for each domain. Again, this process involved team members working toward a consensus. When appropriate, team members could agree to use two words to define a domain (e.g., "adventuresome/risk-taking").

At this point, the lists of domains were regarded as final. The result was a list of features that contained the core ideas about masculinity and femininity held by the persons in our sample. Eight lists were thus formed. We had prototypes of both femininity and masculinity given by younger women, younger men, older women, and older men.


The final rankings of feminine and masculine traits are reported in Tables 1-4. In effect, these lists form the prototypes of femininity and masculinity for the participants in our study. The results for prototypical femininity are presented in Tables 1 and 3 (for younger and older adults, respectively). The findings for prototypical masculinity are displayed in Tables 2 and 4 (for younger and older adults, respectively).

To compare the rankings, a series of Wilcoxen Matched-Pairs tests was conducted. We wanted to identify statistically significant differences in the frequencies of reported features by gender and age. Within both the younger adult and older adult samples, features were matched across female and male respondents within each of the feminine and masculine categories. Analyses were conducted separately for feminine and masculine features.

Three significant differences between pairs of rankings were found (Table 5). First, all younger adults (females and males combined) differed from all older adults in their rankings of masculine traits (z = 2.77, p < .004). Second, younger males differed from older males in their rankings of masculine traits (z = 2.77, p < .006). Third, younger men and younger women differed in their rankings of feminine traits (z = 2.21,p < .03).

Two other trends are worth noting: differences between younger women and older women on feminine traits nearly reached significance (z = 1.94, p < .052); and differences between younger women and older women regarding masculine traits were nearly significant (z = 1.78, p < .07) as well. Other comparisons of ranking pairs were not significant.


Younger and older adults agreed on the top three descriptors of femininity: low self-esteem, emotionally distressed, and restricted/oppressed. Although they differed slightly on the order in which they ranked these three features, both younger and older adults placed them at the center of their femininity prototype. Among younger adults, 32% of all their descriptors fit one of these three categories. In the older adult sample, 29% fit into one of these three groups.

Examples of the language used by participants to describe low self-esteem include "no confidence in herself," "worthless," "insignificant," "feels inferior," "critical of self," and, "unsatisfied with self." Terms that indicated emotional distress included "frustrated," "uptight," "unsure," "lost," "slumping," "troubled," "scared," "confused," "unhappy life," "disillusioned," "depressed," and "suicidal." Phrases that illustrate the restricted/oppressed category include "being controlled," "oppressed," "trapped," "wants out but is stuck," "getting pushed around," "has no control," "confined," and "pushed down."

We used the Chi-square goodness-of-fit test to identify significant differences in matched pairs of rankings. Because of the number of comparisons made, we set Type I error rate at .01. When matched pairs of femininity rankings were evaluated, the only significant difference was between younger men and younger women. At the top of the list, however, they agreed that the central features of femininity are low self-esteem and emotional distress. The two traits that younger women rated higher than younger men revealed a paradox: both strong/determined, [X.sup.2] (1) = 9.93, p <. 01; and passive/dependent, [X.sup.2] (1) = 9.98, p < .01.

In addition, younger women used two descriptors of femininity that younger men completely ignored: self-awareness and self-confidence. Similarly, younger men also used two terms to describe femininity that younger women did not use at all: egocentric and obsessive/compulsive. Otherwise, no statistically significant differences in prototypical femininity were found, either by age or by gender.


Younger and older adults agreed on the core features of masculinity. Although they disagreed slightly on the order in which they were ranked, the top five domains were successful/accomplished, adventuresome/risk-taker, strong/determined, problem solver, and emotionally distressed. These notions form the heart of the masculinity prototype in our sample.

The two most frequently mentioned traits in the prototype were successful/accomplished and adventuresome/risk taker. When participants assumed that a male made the stimulus drawing, they very frequently used descriptors in one of these two categories. In fact, 21% of all the younger adult descriptors and 28% of all the older adult descriptors of masculinity were in one of these two categories. Examples of words that participants used in the "Successful/Accomplishment" category include "ambitious," "career-focused," "driven by accomplishment," "successful," "sets high goals," and "wants to be king of the hill." Examples of descriptors in the "Adventuresome/Risk-taking" group are "daring," "outdoorsy," "fearless," "courageous," "explorer," "risky," and "likes dangerous sports." By contrast, it should be noted that these words were not used when a research participant thought a woman made the very same drawing. In other words, the same drawing elicited very different descriptors, depending solely on the presumed sex of the person who made the drawing.

Generational differences in describing masculinity were significant (z = 2.77, p < .004). Looking more closely at the differences, we identified which traits likely accounted for these generational differences in prototypical masculinity. Two matched pairs of masculinity rankings were statistically significant. The first comparison was between younger adults and older adults. Compared with older adults, younger adults saw prototypical men as experiencing more role strain problems, [X.sup.2] (1) = 68.45, p < .001, and as being more emotionally expressive, [X.sup.2] (1) = 11.08, p < .001. In addition, older adults mentioned four masculinity traits that younger adults did not mention frequently enough to make the list: materialistic, passive/dependent, restricted/oppressed, and aggressive/domineering.

The second significant pair of masculinity rankings was between younger and older men. Younger men mentioned only one masculinity trait more frequently than did older men: intelligent/analytical, [X.sup.2] (1) = 10.62, p < .01.

Younger and older women also ranked several masculinity traits differently. Features that younger women mentioned more frequently were role strain (mentioned 57 times by younger women, not at all mentioned by older women), realistic/pragmatic, [X.sup.2] (1) = 11.31, p < .001; intelligent/analytical, [X.sup.2] (1) = 7.6, p < .01; and optimistic/hopeful, [X.sup.2] (1) = 13.75, p < .001. Older women also used some concepts that younger women did not: passive-dependent, restricted/oppressed, materialistic, and aggressive/domineering.


Three findings stand out in our study. First, there is wide intergenerational agreement on the central features of prototypical femininity and masculinity. Second, as we move away from the center of the prototypes, intergenerational differences become increasingly apparent. Third, the only intragenerational difference involved contrasting views of femininity. All three of these findings warrant further comment.


First, many North Americans still agree that gender differences exist. Although it is often noted that individual females and males are much more alike than different (Maccoby, 1990, 1998), our sample suggests that differences in perception continue and are prominent. At the center of the femininity prototype are low self esteem, restriction/oppression, and emotional distress. Prototypical masculinity emphasizes successful/accomplishment, adventuresome/risk-taking, strength/determination, problem solving, and emotional distress. These differences emerged, even though our method of gathering the data was indirect; we never asked a question about a person's views of either gender. In our manipulation check, only 4.7% of all our participants correctly guessed that the purpose of the study was to explore gender-related constructs. Further, the very same drawings elicited these contrasting descriptions. Participants who assumed that a male made a drawing simply used different words to describe the person than did participants who looked at the very same drawing presuming that a female had drawn it.

Most of these central descriptors have been used to describe women and men for many decades. In fact, they describe much of what is considered traditional gender role behavior. These differences apparently remain, in spite of the national discussion about gender role flexibility that has engaged North Americans over the last 30 years or so.

The one possible surprise near the center of the prototypes is how frequently men were seen as emotionally distressed. In our sample, this distress was described with such words as: "run down," "sad," "hopeless," discouraged," "confused," "lost," "neglected," "restless," "not appreciated," "overwhelmed," "half alive," and "despairing." These descriptors are in direct contrast to the way men have been traditionally socialized (David & Brannon, 1976): being in control, being successful, and never needing help.

Although we do not know with certainty why this feature was so prominent in our masculinity prototype, it is consistent with much theory and research conducted on the experience of men as men in the last 20 years. Several developers of masculinity measures have addressed this theme very directly, describing their theoretical models as "gender-role conflict" (O'Neil, Helms, Gable, David, & Wrightsman, 1986), "sex-role strain" (Pleck, 1981), or "gender-role stress" (Eisler & Skidmore, 1987). These models assert that many men experience high levels of distress when they struggle with the demands of traditional masculinity. In addition, we wonder if some men experience distress when confronted with the idea of changing various gender-related expectations. Our research participants (both old and young) were quite willing to consider emotional distress when they believed that the person who made the drawing was a male.

Our findings also can be compared to the prototypes found by Helgeson in 1994. Although her method of gathering the data was substantially different from ours, several similarities emerged. With regard to the femininity prototype, both her list and ours included caring, self-confidence, a concern with appearance, sociability, friendliness, intelligence, emotionality, and creativity. However, our top three features of low self-esteem, restriction/oppression, and emotional distress did not appear on her list (the one possible exception was "insecure," which appeared on the feminine male list). Perhaps this difference can be explained by the difference in methodology. Helgeson's subjects were asked to define feminine females and males with whatever words came to mind. It may be that impression management prevented her participants from volunteering more negative traits.

Similarly, our masculinity prototype shared several features with Helgeson's prototype. Both her prototype and ours included references to being adventuresome, self-confident, arrogant/egocentric, concerned with physique, intelligent, aggressive/domineering, and persistent or determined. Interestingly, Helgeson's list did not include anything similar to the emotional distress mentioned by our participants, nor any references to masculine role strain. Again, the explanation may lie in difference in methodology. When asked to describe a stimulus person, other factors (social desirability, impression management) may have prevented her participants from volunteering more negative traits. It is also possible that these differences reflect actual trends taking place in North American culture.


The second primary finding is the difference between generations. As we move away from the center of the prototype, these differences become more obvious. This was especially apparent with the masculinity prototype. Younger adults now emphasize the reality of role strain and emotional expressiveness in their understanding of masculinity. These notions are new to masculinity (compare with Parsons, 1955). Our younger participants often speculated about the possibility of gender role strain when they believed the drawing was made by a male: "feeling pressure to be macho," "forced to support family," "intimidated boy," and "confined in his role."

References to male expressivity also were more frequent among young adults than among older adults. Perhaps consistent with this pattern was the fact that younger adults also described men as feeling "alone/shy" more frequently than older adults. This finding should be encouraging to gender theorists who have been calling for the development of a broader array of male traits, especially the ability to be more emotionally expressive (cf., Levant, 1992, 1995).

When compared with younger adults, older adults ranked several masculinity features more highly: appearance/physique, planful, egocentric, materialism, dependency, and aggression. Along with the top five features, these descriptors have been part of the ideology of traditional masculinity for a long time (David & Brannon, 1976; Pleck, 1981). Perhaps it is not surprising that the older adults in our sample focused more on these features. One example illustrates the traditionality of these elements. When older adults commented about male appearance/physique, they generally were not providing a physical description, but rather were describing strength and performance: "muscular," "feels that height is important," "wants to be a good athlete." Younger adults were more likely to refer to visual appearance: "dark hair," "stocky," "bearded," "unattractive," or "plain."

With regard to femininity, it will be disheartening for many feminists to note that after 30 years of pro-feminist influence, younger women indicated that the three central features of prototypical femininity are low self-esteem, emotional distress, and restricted/oppressed. In fact, the picture may be even more striking, in that younger women rated low self-esteem higher than older women did.

Generational differences also were observed with regard to femininity. Traits ranked higher by young adults compared with older adults were: alone/shy, sociable/friendly, intelligent/analytical, independent, and creative. Older adults ranked several traits higher than younger adults ranked them: concern with appearance/ physique, problem solving, obsessive-compulsive, likely to experience role strain, and likely to seek help.

Some of these generational differences do suggest that some changes in gender role for women have taken hold. Younger women see femininity as being more hopeful/optimistic than older women, and more likely to be intelligent/analytical. Older women were more likely to mention features of traditional femininity: help seeking ("cannot succeed alone," "accepting guidance," and "expect help from others"), expressing positive emotions ("pleasant," "happy," and "friendly"), and a concern with appearance/physique ("pretty," "dreamy looking," "attractive," and "stylish").


The only significant within-generation difference involved evaluations of femininity. When matched pairs of femininity rankings were evaluated, the only significant difference was between younger women and younger men. That is to say, younger women and younger men used different descriptors when they thought a female made the drawing.

Although younger women and younger men agreed that the central features of the femininity prototype were low self-esteem and emotional distress, we found that younger men rated the following femininity traits higher than younger women ranked them: help-seeking and alone/shy. On the other hand, the two traits that women rated higher revealed a paradox: their view of femininity was both strong/determined and passive-dependent. In addition, younger women used several descriptors of femininity that younger men virtually ignored: self-awareness, self-confidence, and role strain.

Some of these findings are not encouraging to pro-feminist women and men. When younger men believed a woman made the drawing, they used highly traditional language: low self-esteem, emotional distress, help seeking, and alone/shy. The picture improved only slightly when younger women did the rankings. For them, femininity also included some features that indicate movement away from traditional feminine constructs. They were much more likely than younger men to identify strong/determined, self-awareness, and self-confidence as significant. Not surprisingly, this also created higher levels of gender role strain in younger women.

It would appear that changes away from traditional gender role constructs of femininity are occurring in younger adults, but more so in younger women. The changes are not dramatic, as the younger women in our sample do see prototypical femininity as including strongly traditional traits.

Three limitations to this study need to be noted. The drawings that were used in this study were metaphors created by people to illustrate how they coped with difficulties in their lives. Although we deliberately chose to use these drawings because they offered no overt reference to gender, this stimulus may have limited the range of content. It may be that a wider array of stimuli would have elicited a greater variety of responses, and hence more diversity in the prototypes. Although we selected the drawings because they represented distinctly different topics, we may have suffered from a somewhat restricted range. Nevertheless, the descriptors that did emerge covered a wide range of issues and themes.

A second limitation is the generalizability of these findings, especially from the younger adult group. The younger group consisted of university students at a Midwestern university. Perhaps other young adults would have answered the questions somewhat differently. The older adults, however, shared only their current residency in the Midwest. Other than that, they came from many different organizations and settings.

A third limitation is that other variables besides age might have accounted for the gender differences we found between younger and older adults. For example, these two groups also differed on educational status (younger adults were more likely to be in school) and on relationship status (younger adults were less likely to be married). Perhaps these other variables accounted for some of the variance. It is certainly true that these two variables (and perhaps others that we did not identify) distinguish younger adults from older adults--not only in our sample, but also in society at large. It is useful to keep this in mind when interpreting our findings.

Future research may continue to explore this issue, as social and political forces in North America continue to press for more gender role flexibility. While it may be true that similarities exceed distinctions between women and men, various differences in perception remain and likely shape the way countless women and men interact with each other. The theoretical work exploring these issues will likely continue for some time under various headings (e.g., biological or sociobiological theories, social constructionist views, social role theories). While the qualitative findings of this paper do not speak directly to these theoretical differences, they do present additional evidence that gender differences still exist and have implications for a wide variety of public policy issues.

It will be useful to continue monitoring these patterns with a variety of research methodologies. Further, as researchers give more attention to the individual features that carry more risk for people (such as low self-esteem and emotional distress for women, or role strain and emotional distress for men), preventive and educational programs might be employed to reduce the negative consequences of these tendencies.


Figure 1. Stimulus drawings given to research participants who were asked, "What words would you use to describe the person who made this drawing? You may consider anything that comes to mind--personality traits, interests, physical appearance, typical behavior--whatever seems relevant to you."


We wish to acknowledge the assistance of Sara Hernstrom, Jack Windhorst, Chris Nichols, and Michelle Myers in collecting and analyzing the data.


Allen, B. P. (1995). Gender stereotypes are not accurate: A replication of Martin (1987) using diagnostic vs. self-report and behavioral criteria. Sex Roles, 32(9/10), 583-600.

Bakan, D. (1966). The duality of human existence: An essay on psychology and religion. Chicago: Rand McNally.

Baldwin, A. C., Critelli, J. W., Stevens, L. C., & Russel, S. (1986). Androgyny and sex role measurement: A personal construct approach. Journal of Personality and Social Psychology, 51(5), 1081-1088.

Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162.

Bem, S. L., Martyna, W., & Watson, C. (1976). Sex typing and androgyny: Further explorations of the expressive domain. Journal of Personality and Social Psychology, 34, 1016-1023.

Chesler, P. (1972). Women and madness. Garden City, New York: Doubleday.

Condon, R. G., & Stern, P. R. (1993). Gender-role preference, gender identity, and gender socialization among contemporary Inuit youth. Ethos, 21(4), 384-416.

Constantinople, A. (1973), Masculinity-femininity: An exception to a famous dictum? Psychological Bulletin, 80, 389-407.

Courtenay, W. H. (2000). Engendering health: A social constructionist examination of men's health beliefs and behaviors. Psychology of Men and Masculinity, 1, 4-15.

David, D. S., & Brannon, R. (Eds.). (1976). The forty-nine percent majority: The male sex role. New York: Random.

Eisler, R. M., & Skidmore, J. R. (1987). Masculine gender role stress: Scale development and component factors in the appraisal of stressful situations. Behavior Modification, 11(2), 123-136.

Ember, M., & Ember, C. (1971). The conditions favoring matrilocal vs. patrilocal residence. American Anthropologist, 73, 571-594.

Frank, S. J., McLaughlin, A. M., & Crusco, A. (1984). Sex role attributes, symptom distress, and defensive style among college men and women. Journal of Personality and Social Psychology, 47, 182-192.

Gelso, C. (1979). Research in counseling: methodological and professional issues. Counseling Psychologist, 8(3), 7-35.

Gergen, K. (1991). The saturated self. New York: Basic Books.

Gilligan, C. (1982). In a different voice: Psychological theory and women's development. Cambridge, MA: Harvard University Press.

Goldberg, H. (1976). The hazards of being male: Surviving the myth of masculine privilege. New York: Signet.

Hathaway, S. R., & McKinley, J. C. (1940). A multiphasic personality schedule (Minnesota): Construction of a schedule. Journal of Psychology, 10, 249-254.

Helgeson, V. S. (1990). The role of masculinity in a prognostic predictor of heart attack severity. Sex Roles, 22, 755-774.

Helgeson, V. S. (1994). Prototypes and dimensions of masculinity and femininity. Sex Roles, 31(11/12), 653-682.

Highet, A. (1998). Casablanca, Humphrey Bogart, the Oedipus complex, and the American male. Psychoanalytic Review, 85(5), 761-774.

Hill, C. E., Nutt-Williams, E., Heaton, K. J., Thompson, B. J., & Rhodes, R. H. (1996). Therapist retrospective recall of impasses in long-term psychotherapy: A qualitative analysis. Journal of Counseling Psychology, 43, 207-217.

Hill, C. E., Thompson, B. J., & Williams, E. N. (1997). A guide to conducting Consensual Qualitative Research. The Counseling Psychologist, 25(4), 517-572.

Holmes, C. B., & Wiederholt, J. (1982). Depression and figure size on the Draw-A-Person Test. Perceptual and Motor Skills, 55, 825-826.

Hort, B. E., Fagot, B. I., & Leinbach, M. D. (1990). Are people's notion of maleness more stereotypically framed than their notion of femaleness?" Sex Roles, 23, 197-212.

Houston, A. N., & Terwilliger, R. (1995). Sex, sex roles, and sexual attitudes: Figure gender in the Draw-A-Person Test revisited. Journal of Personality Assessment, 65(2), 343-357.

Johnson, A. (1997). Narratives and metaphors of college students identified as learning disabled. Unpublished doctoral dissertation, Antioch New England Graduate School, Keene, New Hampshire.

Jones, E. E., & Nisbett, R. E. (1972). The actor and the observer: Divergent perceptions of the causes of behavior. In E. E. Jones et al. (Eds.), Attribution: Perceiving the causes of behavior (pp. 79-94). Morristown, NJ: General Learning Press.

Judd, C. M., & Park, B. (1993). Definition and assessment of accuracy in social stereotypes. Psychological Review, 10, 109-128.

Karbon, M., Fabes, R. A., Carlo, G., & Martin, C. L. (1992). Preschoolers' beliefs about sex and age differences in emotionality. Sex Roles, 27(7/8), 377-390.

Kelly, G. A. (1955). The psychology of personal constructs. New York: Norton.

Knox, S., Hess, S. A., Petersen, D. A., & Hill, C. E. (1997). A qualitative analysis of client perceptions of the effects of helpful therapist self-disclosure in long-term therapy. Journal of Counseling Psychology, 44, 274-283.

Ladany, N., O'Brien, K. M., Hill, C. E., Melincoff, D. S., Knox, S., & Petersen, D. A. (1997). Sexual attraction toward clients, use of supervision, and prior training: A qualitative study of psychotherapy predoctoral interns. Journal of Counseling Psychology, 44(4), 413-424.

Levant, R. (1992). Toward the reconstruction of masculinity. Journal of Family Psychology, May/June, 379-402.

Levant, R. (1995). Masculinity reconstructed. New York: Dutton.

Lilienfeld, S. O., Wood, J. M., & Garb, H. N. (2000). The scientific status of projective techniques. Psychological Science in the Public Interest, 1(2), 27-65.

Locksley, A. & Colten, M. E. (1979). Psychological androgyny: A case of mistaken identity? Journal of Personality and Social Psychology, 37, 1017-1031.

Lott, B. (1979). Sex-role ideology and children's drawings: Does the jack-o-lantern smile or scare? Sex Roles, 5(1), 93-98.

Luca-Stolkin, M. (1999). Pandora's box: The shadow of femininity in the treatment of psycho-somatic distress. Psychodynamic Counselling, 5(2), 173-191.

Maccoby, E. (1990). Gender and relationships: A developmental account. American Psychologist, 45(4), 513-520.

Maccoby, E. (1998). The two sexes: Growing up apart, coming together. Cambridge, MA: Belknap Press/Harvard University Press.

Machover, K. (1949). Personality projection in the drawing of the human figure. Springfield, IL: Thomas.

Markus, H., & Oyserman, D. (1989). Gender and thought: The role of the self-concept. In M. Crawford & M. Gentry (Eds.), Gender and thought: Psychological perspectives (pp. 100-127). New York: Springer-Verlag.

Martin, C. L. (1987). A ratio measure of gender stereotyping. Journal of Personality and Social Psychology, 52, 489-499.

Munroe, R., Munroe, R., & Whiting, J. (1981). Male sex-role resolutions. In R. Munroe, R. Munroe, & B. Whiting, (Eds.), Handbook of cross-cultural human development (pp. 611-632). New York: Garland Press.

Murdock, G.P., & Provost, C. (1973). Factors in the sex division of labor. Ethnology, 12, 203-225.

Myers, A. M,, & Gonda, G. (1982). Utility of the masculinity-femininity construct: Comparison of traditional and androgyny approaches. Journal of Personality and Social Psychology, 43, 514-523.

O'Neil, J. M., Helms, B., Gable, R., David, L., & Wrightsman, L. (1986). Gender Role Conflict Scale: College men's fear of femininity. Sex Roles, 14, 335-350.

Ortega, D., Arnold, L. E., Smeltzer, D. J. (1988). Analysis of children's human figure drawings. Journal of the American Academy of Child and Adolescent Psychiatry, 27(2), 260, 261.

O'Toole, J. A., Defenbacher, K. A., Valentin, O., McKee, K., Huff, D., & Abdi, H. (1998). The perception of gender: The role of stimulus structure in recognition and classification. Memory and Cognition, 26(1), 146-160.

Parsons, T., & Bales, R. F. (1955). Family, socialization and interaction processes. New York: The Free Press.

Pennell, G. E., & Ogilvie, D. M. (1995). You and me as she and he: The meaning of gender-related concepts in other- and self-perception. Sex Roles, 33 (1/2), 29-57.

Philpot, C. L. (2000). Gender stereotyping has diminished but society still operates with "old lenses." The National Psychologist, July/August, p. 14.

Pleck, J. H. (1981). The myth of masculinity. Cambridge, MA: MIT Press.

Raty, H., & Snellman, L. (1997). Children's images of an intelligent person. Journal of Social Behavior and Personality, 12(3), 773-784.

Rubenstein, J., Feldman, S. S., Rubin, C., & Noveck, I. (1987). A cross-cultural comparison of children's drawings of same- and mixed-sex peer interaction. Journal of Cross-cultural Psychology, 18(2), 234-250.

Silver, R. A. (1987). Sex differences in the emotional content of drawings. Art Therapy: Journal of the American Art Therapy Association, 4(2), 67-77.

Snarey, J., & Son, L. (1986). Sex-identity development among kibbutz-born males: A test of the Whiting Hypothesis. Ethos 14(2), 99-119.

Spence, J. T. (1984). Masculinity, femininity, and gender-related traits: A conceptual analysis and critique of current research. In B. A. Maher & W. B. Maher (Eds.), Progress in experimental personality research (Volume 13) (pp. 2-97). Orlando, FL: Academic Press.

Spence, J. T., & Helmreich, R. L. (1978). Masculinity & femininity: Their psychological dimensions, correlates, and antecedents. Austin, TX: University of Texas Press.

Spence, J. T., Helmreich, R. L., & Holahan, C. K. (1979). Negative and positive components of psychological masculinity and femininity and their relationship to self-reports of neurotic and acting out behavior. Journal of Personality and Social Psychology, 37, 1673-1682.

Spence, J. T., Helmreich, R. L., & Stapp, J. (1974). The Personal Attributes Questionnaire: A measure of sex-role stereotypes and masculinity-femininity. JSAS Catalogue of Selected Documents in Psychology, 4, 127.

Strong, E. K. (1936). Interests of men and women. Journal of Social Psychology, 7, 49-67.

Styles, W. B. (1993). Quality control in qualitative research. Clinical Psychology Review, 13, 593-618.

Swim, J. K. (1994). Perceived versus meta-analytic effects sizes: An assessment of the accuracy of gender stereotypes. Journal of Personality and Social Psychology, 66, 21-36.

Tavris, C., & Wade, D. (1984). The longest war: Sex differences in perspective. San Diego, CA: Harcourt Brace Jovanovich.

Terman, L., & Miles, C. C. (1936). Sex and personality, studies in masculinity and femininity. New York: McGraw-Hill.

Twenge, J. M. (1999). Mapping gender: The multifactorial approach and the organization of gender-related attributes. Psychology of Women Quarterly, 23, 485-502.

Wheeler, L. Reis, H., & Nezlek, J. B. (1983). Loneliness, social interaction, and sex roles. Journal of Personality & Social Psychology, 45(4), 943-953.

White, H., Morris, A., & Arthur, A. G. (1996). Gender role stereotyping in picture interpretation. The Journal of Genetic Psychology, 157(1), 119-120.

Correspondence concerning this article should be addressed to John M. Robertson, University Counseling Services, Kansas State University, Manhattan, Kansas 66506. Electronic mail may be sent to johnrobe@ksu.edu.
Table 1

Prototypical Features of Femininity: Young Adult Rankings

Feature                        Combined Ranking
                                  (N = 192)

                             Rank      Frequency

Low Self-Esteem                1           135
Emotionally Distressed         2           109
Restricted/Oppressed           3            84
Alone/Shy                      4            69
Successful/Accomplished        5            67
Sociable/Friendly              6            61
Strong/Determined              7            58
Passive/Dependent              8            53
Optimistic/Hopeful             9            50
Emotionally Expressive        10            47
Intelligent/Analytical        11            45
Appearance/Physique           12            40
Help-Seeking                  13            39
Problem Solver                14            36
Care-giving/Empathic          15            28
Independent                   16            23
Self-Awareness                17            17
Creative                      18            13
Self-Confident                19            12
Egocentric                    20            11
Obsessive/Compulsive          21             9
Positive Affect               22             7
Realistic/Pragmatic           23             4
Role Strain                   24             1

Feature                         Male Ranking
                                 (N = 71)

                             Rank      Frequency

Low Self-Esteem                1            59
Emotionally Distressed         2            57
Restricted/Oppressed           4            37
Alone/Shy                      3            40
Successful/Accomplished        5            30
Sociable/Friendly              6            22
Strong/Determined             12            17
Passive/Dependent            13t            15
Optimistic/Hopeful             9            20
Emotionally Expressive       10t            18
Intelligent/Analytical        7t            21
Appearance/Physique          10t            18
Help-Seeking                  7t            21
Problem Solver               13t            15
Care-giving/Empathic         13t            15
Independent                   18             7
Self-Awareness                nr             0
Creative                      19             5
Self-Confident                nr             0
Egocentric                    16            11
Obsessive/Compulsive          17             9
Positive Affect              20t             4
Realistic/Pragmatic          20t             4
Role Strain                   nr             0

Feature                        Female Ranking
                                  (N = 121)

                             Rank       Frequency

Low Self-Esteem                1            76
Emotionally Distressed         2            52
Restricted/Oppressed           3            47
Alone/Shy                     9t            29
Successful/Accomplished        7            37
Sociable/Friendly              5            39
Strong/Determined              4            41
Passive/Dependent              6            38
Optimistic/Hopeful             8            30
Emotionally Expressive        9t            29
Intelligent/Analytical        11            24
Appearance/Physique           12            22
Help-Seeking                  14            18
Problem Solver                13            21
Care-giving/Empathic          17            13
Independent                   16            16
Self-Awareness                15            17
Creative                      19             8
Self-Confident                18            12
Egocentric                    nr             0
Obsessive/Compulsive          nr             0
Positive Affect               20             3
Realistic/Pragmatic           nr             0
Role Strain                   21             1

Note. Features are rank ordered by frequency count. nr = not
ranked. Ties are signified by a "t."

Table 2

Prototypical Features of Masculinity: Young Adult Rankings

Feature                         Combined Ranking
                                   (N = 192)

                              Rank       Frequency

Successful/Accomplished         1           193
Adventuresome/Risk-Taker        2           150
Strong/Determined               3           138
Emotionally Distressed          4           114
Problem Solver                  5           103
Alone/Shy                       6           102
Low Self-Esteem                 7            94
Self-Confident                  8            85
Independent                     9            79
Role Strain                    10            77
Intelligent/Analytical         11            76
Optimistic/Hopeful             12            71
Planful                        13            60
Appearance/Physique            14            59
Positive Affect                15            47
Realistic/Pragmatic            16            40
Emotionally Expressive        17t            38
Help-Seeking                  17t            38
Competitive                    19            33
Sociable/Friendly              20            26
Egocentric                     21            22
Creative                       22            16

Feature                         Male Ranking
                                  (N = 71)

                              Rank      Frequency

Successful/Accomplished         1           100
Adventuresome/Risk-Taker        3            82
Strong/Determined               2            83
Emotionally Distressed         10            42
Problem Solver                 12            37
Alone/Shy                       5            49
Low Self-Esteem                6t            47
Self-Confident                 8t            46
Independent                    8t            46
Role Strain                    16            20
Intelligent/Analytical          4            53
Optimistic/Hopeful            13t            31
Planful                        6t            47
Appearance/Physique            11            41
Positive Affect               13t            31
Realistic/Pragmatic           20t            10
Emotionally Expressive        20t            10
Help-Seeking                   15            20
Competitive                    17            18
Sociable/Friendly              18            16
Egocentric                     19            12
Creative                       22             6

Feature                         Female Ranking
                                   (N = 121)

                              Rank       Frequency

Successful/Accomplished         1            93
Adventuresome/Risk-Taker        3            68
Strong/Determined               6            55
Emotionally Distressed          2            72
Problem Solver                  4            66
Alone/Shy                       7            53
Low Self-Esteem                 8            47
Self-Confident                 10            39
Independent                    11            33
Role Strain                     5            57
Intelligent/Analytical         13            23
Optimistic/Hopeful              9            40
Planful                       18t            13
Appearance/Physique           14t            18
Positive Affect                17            16
Realistic/Pragmatic            12            30
Emotionally Expressive        14t            18
Help-Seeking                  14t            18
Competitive                   18t            13
Sociable/Friendly             20t            10
Egocentric                    20t            10
Creative                      20t            l0

Note. Features are rank ordered by frequency count. Ties are signified
by a "t."

Table 3

Prototypical Features of Femininity: Older Adult Rankings

Feature                        Combined Ranking
                                  (N = 126)

                             Rank      Frequency

Emotionally Distressed         1            91
Restricted/Oppressed           2            90
Low Self-Esteem                3            71
Appearance/Physique            4            56
Passive/Dependent              5            53
Problem Solver                 6            50
Successful/Accomplished        7            49
Emotionally Expressive         8            46
Help-Seeking                   9            45
Strong/Determined             10            44
Positive Affect               11            35
Sociable/Friendly             12            33
Optimistic/Hopeful            13            31
Obsessive/Compulsive          14            29
Self-Confident                15            25
Intelligent/Analytical        16            24
Alone/Shy                     17            22
Role Strain                   18            15
Planful                       19            13
Realistic/Pragmatic           20            11
Independent                   21            10
Egocentric                    22             9
Creative                      23             6

Feature                          Male Ranking
                                   (N = 58)

                             Rank      Frequency

Emotionally Distressed         2            42
Restricted/Oppressed           5            30
Low Self-Esteem                1            51
Appearance/Physique            4            34
Passive/Dependent              3            38
Problem Solver                 8            22
Successful/Accomplished       11            15
Emotionally Expressive         7            25
Help-Seeking                  10            16
Strong/Determined            12t            13
Positive Affect                9            18
Sociable/Friendly            15t            12
Optimistic/Hopeful           12t            13
Obsessive/Compulsive           6            29
Self-Confident               17t            11
Intelligent/Analytical       17t            11
Alone/Shy                     nr             0
Role Strain                  15t            12
Planful                      12t            13
Realistic/Pragmatic          17t            11
Independent                   21             1
Egocentric                    20             9
Creative                      nr             0

Feature                         Female Ranking
                                   (N = 68)

                             Rank       Frequency

Emotionally Distressed         2            49
Restricted/Oppressed           1            60
Low Self-Esteem               11            20
Appearance/Physique           7t            22
Passive/Dependent             14            15
Problem Solver                 6            28
Successful/Accomplished        3            34
Emotionally Expressive        9t            21
Help-Seeking                   5            29
Strong/Determined              4            31
Positive Affect               13            17
Sociable/Friendly             9t            21
Optimistic/Hopeful            12            18
Obsessive/Compulsive          nr             0
Self-Confident                15            14
Intelligent/Analytical        16            13
Alone/Shy                     7t            22
Role Strain                   19             3
Planful                       nr             0
Realistic/Pragmatic           nr             0
Independent                   17             9
Egocentric                    nr             0
Creative                      18             6

Note. Features are rank ordered by frequency count. nr =
not ranked. Ties are signified by a "t."

Table 4

Prototypical Features of Masculinity: Older Adult Rankings

Feature                          Combined Ranking
                                    (N = 126)

                              Rank       Frequency

Successful/Accomplished         1           225
Adventuresome/Risk-Taker        2           103
Problem Solver                  3            96
Emotionally Distressed          4            95
Strong/Determined               5            89
Appearance/Physique             6            62
Self Confident                  7            61
Low Self-Esteem                 8            51
Planful                         9             4
Alone/Shy                      10            42
Independent                    11            36
Realistic/Pragmatic            12            34
Optimistic/Hopeful             13            27
Materialistic                  14            24
Intelligent/Analytical         15            23
Egocentric                    16t            22
Help-Seeking                  16t            22
Positive Affect               16t            22
Passive/Dependent              19            20
Competitive                    20            17
Sociable/Friendly             21t            16
Restricted/Oppressed          21t            15
Emotional Expressiveness       23            14
Aggressive/Domineering         24             9
Creative                       25             8
Role Strain                    26             3

Feature                           Male Ranking
                                    (N = 58)

                              Rank       Frequency

Successful/Accomplished         1           104
Adventuresome/Risk-Taker        6            39
Problem Solver                  2            54
Emotionally Distressed          3            47
Strong/Determined               4            46
Appearance/Physique             7            30
Self Confident                  5            45
Low Self-Esteem                16            12
Planful                        9t            21
Alone/Shy                      9t            21
Independent                    11            19
Realistic/Pragmatic             8            25
Optimistic/Hopeful             15            14
Materialistic                  19            10
Intelligent/Analytical         14            15
Egocentric                    17t            11
Help-Seeking                  17t            11
Positive Affect                13            16
Passive/Dependent              nr             0
Competitive                    12            17
Sociable/Friendly             20t             4
Restricted/Oppressed           nr             0
Emotional Expressiveness      20t             4
Aggressive/Domineering         nr             0
Creative                       23             2
Role Strain                    22             3

Feature                          Female Ranking
                                   (N = 68)

                              Rank       Frequency

Successful/Accomplished         1           121
Adventuresome/Risk-Taker        2            64
Problem Solver                  5            42
Emotionally Distressed          3            48
Strong/Determined               4            43
Appearance/Physique             7            32
Self Confident                 11            16
Low Self-Esteem                 6            39
Planful                         8            24
Alone/Shy                       9            21
Independent                    11            17
Realistic/Pragmatic            21             9
Optimistic/Hopeful             15            13
Materialistic                  14            14
Intelligent/Analytical         22             8
Egocentric                    17t            11
Help-Seeking                  17t            11
Positive Affect               23t             6
Passive/Dependent              10            20
Competitive                    nr             0
Sociable/Friendly              16            12
Restricted/Oppressed           13            15
Emotional Expressiveness       19            10
Aggressive/Domineering         20             9
Creative                      23t             6
Role Strain                    nr             0

Note. Features are rank ordered by frequency count. nr = not
ranked. Ties are signified by a "t."

Table 5

Comparison of Masculinity and Femininity Rankings, by Matched Pairs

Comparison                          Male Rankings   Female Rankings

                                    z statistic     z statistic

Younger Adults vs. Older Adults         2.85 **          1.36
Younger Men vs. Older Men               2.77 **           .06
Younger Women vs. Older Women           1.78             1.94
Younger Men vs. Younger Women            .66             2.21 *
Older Men vs. Older Women                .68              .14
All Men vs. All Women                    .25             1.49

* p < .05 ** p < .01
Gale Copyright: Copyright 2002 Gale, Cengage Learning. All rights reserved.