This is an open access article distributed under the terms of the Creative Commons
Attribution License (There is a dilemma in the very nature of research on individual differences. On the one hand, we want to discover the characteristics of a subgroup of people in order to understand them and gain insights regarding the implications of their differences. On the other hand, the attempt to identify any group’s characteristics leads to stereotyping. This dilemma seems even more striking when the sub-grouping of interest, gender, involves groups of people representing roughly 50% of the population each. There are bound to be within-group differences and the dangers of stereotyping create problems in work situations and in relationships in general.
The literature tends to describe females as nurturing, emotional, sensitive, and more aesthetically oriented than males (Brody, 1997). While this tendency may be present, there are many, no doubt millions, of females who do not portray these characteristics. Maybe we need to refine our understanding of gender differences by looking at the personality differences within each gender’s population. Personality traits should have long-term influence on outcomes and, therefore, should be preferred predictors (Phillips & Lloyd, 2006). Specifically, with respect to gender, important variance may be lost in the dichotomous identification of individuals as male or female.
Gardenswartz and Rowe (1994) identified layers of diversity, suggesting that gender is a dimension that helps defines one’s identity, but personality is a deeper level of diversity, closer to the centre of what defines who we are. Personality underlies and offers a more complex explanation for behaviour compared to gender alone. If we really want to understand women (or men), therefore, we need to look at the interaction between gender and personality (Visser & Roelofs, 2011). Despite the identification of gender differences in a variety of work-related behaviours, interactions with deeper-level variables received little attention.
We chose to investigate the happiness construct due to its significant, positive effects on various work-related outcomes and the potential to clarify mixed findings in past research (Bowling, Eschleman, & Wang, 2010; Kroll, 2011). The goal of our study was to investigate whether personality and gender interact in predicting the choice to pursue various happiness strategies, and if this interrelationship is a predictor of strategy choice versus gender or personality as sole predictors. Relevant literature is presented, followed by a research model and hypotheses. Next, method and data are described, along with a discussion of results, limitations and suggestions for future research and application.
Hypothesis Development
Deep- and Surface-Level Diversity
Gardenswartz and Rowe (1994) identified four layers of diversity to help distinguish how people differ. The outer two layers of their model describe externally established dimensions (e.g., organisational field of work, seniority level, marital status, religious affiliation, etc.). Gender is an internal dimension, as is race, and thus closer to a core element of one’s identity. Personality is at the centre of the diversity levels and makes us who we are.
Other researchers have adopted the perspective of diversity ‘layers’ (see Figure 1). For example, researchers labelled such observable traits as gender as sources of “surface-level” diversity (Harrison, Price, & Bell, 1998). Such unobservable types of diversity as personality have been called “deep-level” diversity traits. Surface-level and deep-level traits both influence behaviour (Phillips & Lloyd, 2006); however, deep-level traits become better predictors of outcomes over time, while surface-level traits become less important (Harrison et al., 1998). Perhaps the interaction of the key surface trait of gender and a key personality trait will help better explain perceived differences between men and women.
Figure 1
Surface-level versus deep-level diversity.
Therefore, surface-level characteristics are interpreted as proxies for deep-level characteristics (Chatman, Polzer, Barsade, & Neale, 1998; Chen & Kenrick, 2002; Lawrence, 1997; Phillips, 2003). People may assume deep-level diversity is present when only surface-level diversity actually exists, or vice-versa (Phillips & Lloyd, 2006). This misunderstanding may prohibit the opportunity for deep-level understanding of relationships among the variables (Harrison et al., 1998). If a link between surface- and deep-level diversity can be identified by researchers, then the effects of deep-level traits are more lasting and have longer-term influence, so they would be preferred as predictors.
In summary, researchers agree that personality traits may offer an underlying explanation for gender differences. Identifying the specific personality traits and their links to gender could lead to more lasting, detailed understandings of behaviour. These links may be interactive and could reveal underlying meaning that has been masked by studying only main effects.
Research Findings Relating Gender and Personality
Researchers have investigated the interplay between gender and personality in predicting various behaviours such as in interpersonal communication (Baker & McNulty, 2011); altruism (Ben-Ner, Kong, & Putterman, 2004; Swope, Cadigan, Schmitt, & Shupp, 2008; Visser & Roelofs, 2011); job performance (Saad & Sackett, 2002); entrepreneurship (Zhang et al., 2009); work-life balance (George, Helson, & John, 2011) and happiness (Warner & Vroman, 2011).
Researchers studied the interaction between gender and outcome variables that are stereotypically male or female, such as altruism and communication style. Results have revealed that personality traits explain more variance in addition to gender. Although gender may have predicted the behaviour in a manner consistent with stereotypes, personality interacted with gender to provide more fine-tuned predictions. For example, Visser and Roelofs (2011) investigated the reasons for gender effects on altruistic behaviour. Gender differences in altruism were found only in those individuals whose scores were higher or lower than the average on each personality factor of the Big Five personality traits of conscientiousness, agreeableness, neuroticism, openness, and extraversion (Costa & McCrae, 1992).
Ben-Ner et al. (2004) also investigated the joint effects of gender and personality on the relationship between awareness of gender of a fictitious endowment recipient and altruism toward that person. For women, the relationship between recipient gender and altruism related to each of the Big Five personality factors. Finally, Baker and McNulty (2011) investigated effect of the interaction between gender, self-compassion and conscientiousness on relationship satisfaction and willingness to correct interpersonal errors. Women’s self-compassion positively affected their desires to increase relationship satisfaction and to correct their interpersonal mistakes. These relationships only held for men with high conscientiousness, however.
Other researchers found as well that personality added explanation power to gender when predicting non-stereotypical behaviours. For example, George et al. (2011) found that personality was an important and stable predictor of women’s work behaviours over a 50-year period. Personality traits explained various outcomes such as career choice, work involvement, status, satisfaction, and financial security. Saad and Sackett (2002) hypothesised an interaction between personality and gender on U.S. Army employees’ job performance. Results reveal that job performance was over-predicted for women when personality interactions were included.
In addition, Warner and Vroman (2011) found relationships between happiness-inducing strategies, perceived happiness, gender, and the Big Five personality traits. Women reported significantly higher perceived happiness than men, and significantly higher use of some happiness-inducing strategies. Big Five personality traits also significantly predicted happiness-inducing strategies with conscientiousness, agreeableness, openness, and extraversion as positive predictors. The relationship between happiness-inducing behaviours and perceived happiness was not moderated by gender or by the Big Five traits. The authors note, however, that the study was exploratory and that they used a very short (two items per trait) version of the Big Five measure (Goldberg et al., 2006) and a new measure of perceived happiness. The large sample size (n = 903) may have also led to misleading significant relationships with weak correlations.
Genetic studies added credibility to the knowledge that gender and personality interact to predict behaviour. In a study of twins, Zhang et al. (2009) investigated the relative influence of gender, Big Five personality traits, and environment on the decision to become an entrepreneur. For women, the relationship between genetics and entrepreneurship was mediated by extraversion and neuroticism. For men, the relationship between shared-environmental influences and entrepreneurship was mediated by extraversion.
In summary, researchers have shown that personality and gender jointly predict various work- and relationship-related behaviours. This research spans a wide range of personality tests and a wide range of behavioural outcomes. As a result of this variation, although some indication of a consistent personality-gender link is evident, it is difficult to draw overarching conclusions, given the challenges inherent in comparing studies (Visser & Roelofs, 2011).
Happiness, Gender, and Personality
Happiness research is on the forefront of the positive psychology movement and has recently received increased consideration in organisational research (Judge & Kammeyer-Mueller, 2011). Happiness is an affective component of subjective well-being (SWB) for which researchers have identified “considerable” gender differences in past studies (Kroll, 2011). In addition, Bowling et al. (2010) identified the presence of moderators in their meta-analysis and suggested that gender should be investigated in more detail in future research.
Happiness is positively related to various work-related outcomes, such as job satisfaction (Bowling et al., 2010); and positive moods, motivation, and subjective well-being (Judge & Kammeyer-Mueller, 2011). In addition, happiness is negatively related to depression and anxiety (Borooah, 2010). Studies have proposed happiness ties to organisational outcomes, as well, such as social capital (Leung, Kier, Fung, Fung, & Sproule, 2011); organisational citizenship behaviour (Rego, Ribeiro, & Cunha, 2010); flexible work environments (Atkinson & Hall, 2011); job performance, and organisational functioning (Fredrickson, 2003).
Research supports that happiness can be increased and environmental influence can play a critical role (Nes, 2010). Specifically, behavioural strategies can be used to increase happiness (Warner & Vroman, 2011). In one study, these behavioural strategies appear to vary according to gender and personality, but only as linear predictors and not as interacting variables (e.g., Tkach & Lyubomirsky, 2006). In one study, researchers found a gender-personality interaction in use of happiness strategies, but using exploratory scales (Warner & Vroman, 2011).
In happiness research, strategy choice is typically measured when participants rate their use of various methods that have been shown to increase happiness, including such behaviours as expressing gratitude and active leisure behaviours (Lyubomirsky, 2008). The measurement rating gives the individual a choice of five motivations for a person to persist in performing different activities (i.e., the behaviours are natural to them, they enjoy performing the behaviours, they appreciate the value of the behaviours, or they engage in them out of feelings of guilt or due to situational demands).
Men and women may tend to choose happiness strategies that are considered to be stereotypically gender-appropriate. In fact, individual choices are affected by sociocultural stereotypes (Carr & Steele, 2010). Women, in particular, are more positively evaluated for behaviour that is consistent with social expectations than men (Shaughnessy, Treadway, Breland, Williams, & Brouer, 2011), and therefore may choose ‘feminine’ strategies to induce happiness. Individuals of both genders could even adopt stereotypical methods for gaining happiness, even if their genuine interests and desires are not consistent with these stereotypes. Stereotypes that are widely accepted (but not necessarily true) and could influence happiness strategy choice include the idea that compared to men, women are more emotional (Heesacker et al., 1999) and express different emotions (Plant, Hyde, Keltner, & Devine, 2000). Compared to men, women are also stereotypically thought to be more fearful (Condry & Condry, 1976) and to express more intense embarrassment (Keltner, 1995).
In sum, happiness is an appropriate outcome variable in a study of gender and personality effects. The study of happiness involves behaviour, in the form of strategies used to increase happiness, and affect toward the particular strategies chosen.
The Myers-Brigg Type Indicator (MBTI) and Suitability for Gender Studies
One of the most commonly used personality indicators in business settings is the Myers-Briggs Type Indicator. Although scored in an ipsative fashion (two acceptable options are provided and the participant is forced to choose only one), the scales of the MBTI have been repeatedly found to be of high reliability with abundant evidence of their validity (e.g., Harvey, 1996; Johnson, 1992; Myers, McCaulley, Quenk, & Hammer, 1998; McCrae & Costa, 1989).
Those that have criticized the reliability of the MBTI (Clark & Estes, 2002; Strickler & Ross, 1962) were examining older versions, and the present study used the recently improved version (Form M) and a “best-fit” procedure. Analysis of a national sample data (Myers et al., 1998) showed internal consistency and test-retest reliabilities of the MBTI Form M assessment to be good or superior to those of such commonly used personality measures as the NEO Pi, Birkman Method, DiSC, Bar-On EQ-I and the 16PF. In addition, the Cronbach Alpha measures of internal consistency reliability were found to be between .86 to .92 for all the scales in samples comprising different ethnic and age groups. While considerable research has been done on the MBTI assessment in a variety of countries (Beuke, Freeman, & Wang, 2006; Schaubhut, 2008; Yiannakis & Taylor, 2009), data gathered from Africa, Asia, Australia, Europe, Latin America and the Middle East also showed good evidence of internal consistency / reliability, ranging from .81 to .91 with a pattern similar across all six regions (Schaubhut, Herk, & Thompson, 2009).
Convergent and divergent validity evidence is reported based on correlation patterns with other personality assessments, namely the CPI, FIRO-B, and The Adjective Checklist. The strong correlations are comparable to those found when Big Five factors were measured by the NEO-PI (Myers et al., 1998). The MBTI is based on Jung’s Theory of Personality and has a scale that corresponds to each of the dimensions of the Big Five.1 The results of each scale are reported as dichotomous type preferences (akin to right-handedness versus left-handedness) rather than scores on a trait, but with acknowledgement that behaviourally we may use “either side of the trait” in order to match the demands of a given situation.
In the “best-fit” or “verified type” procedure used to feedback MBTI results, participants are given detailed information on their reported type and alternative type descriptions. A number of studies (Hammer & Yeakley, 1987; Herk & Thompson, 2009) revealed high rates of agreement between type as reported by the instrument and best-fit type (72.9% agree on all 4 scales of the MBTI assessment). We will be using best-fit type in this study in order to ensure that the participants are convinced by the description of their personalities. Studies conducting confirmatory factor analyses (Harvey, Murry, & Stamoulis, 1995; Johnson & Saunders, 1990) provide further evidence of the validity of the MBTI scales.
Study after study has found that females are much more likely to be the “Feeling” personality types and males are much more likely to be “Thinking” types. For example, in the nationally representative sample utilized in the Hammer and Mitchell (1996) study, 75.5% of females’ scores classified them as “Feeling” preference and 24.5% as “Thinking” preference. Among males, 56.5% scored a preference for “Thinking” and 43.5% scored a preference for “Feeling.” Jung used the terms “Thinking” and “Feeling” to describe the two types of rational functions that are used to appraise or evaluate perceptions. Thinking types come to decisions through logical connections relying on the principles of cause and effect. Feeling types come to decisions by weighing the relative values and merits of the issues, relying on an understanding of personal values and aesthetics. Neither approach is inherently superior. The rational judging function scale significantly correlates with Big Five measures of Agreeableness (McCrae & Costa, 1989).
One inherent difficulty in investigating the interactions between gender and personality lies in the trait approach, or in classifying individuals as “high” or “low” on given dimensions. For example, if women are described as more “agreeable” than men, does this mean that men should be described as “disagreeable” (Costa, Terracciano, & McCrae, 2001)?
We suggest that research investigating the interaction effects of gender and personality is better served by using this typological approach. Instead of suggesting that females are more agreeable and males more disagreeable, we propose to investigate gender interactions with the MBTI thinking and feeling dimensions. The preference for Thinking over Feeling in the MBTI framework indicates individuals who are more willing to analyse and critique ideas. Thus we are proposing to study four MBTI types: Feeling Females, Feeling Males, Thinking Females and Thinking Males.
Based on the above literature, we conclude that personality is a deep-level diversity trait and gender is a surface-level diversity trait. We conclude also that gender differences in work-related behaviours may be better understood by their interaction with personality traits, especially those that classify individuals into ‘types’, allowing for a better understanding of each group rather than a “high” versus “low” on a single trait. Last, we propose that happiness is a valuable construct for studying personality-gender interaction effects. Because happiness is studied in terms of behavioural choices of strategies as well as affect for those strategies, a multifaceted understanding may be gained that gives more information about gender-personality relationship as predictors. Research regarding choice behaviour suggests that sociocultural stereotypes may affect evaluation of the happiness strategies one uses, but not enough evidence is available to specifically hypothesize results. Therefore, based on the literature presented here, we offer the following hypotheses:
H1: Choice of happiness strategies will vary significantly according to participant gender.
H2: Choice of happiness strategies will vary significantly according to participant personality.
H3: Choice of happiness strategies will vary significantly according to an interaction between participant gender and participant personality. Specifically, Females with “Feeling” MBTI classification and Males with the “Thinking” classification are expected to exhibit the largest number of significant differences in strategy choices.
Method
Participants
The sample of this study included 208 students in the graduate and undergraduate sections of a course on organisational behaviour. Four surveys were not completed and, therefore, were not included in the sample. The usable sample included 103 females (50.50%) and 101 males (49.50%). The sample was composed of 98 students who were identified as Thinking types (48.04%) and 106 who were identified as Feeling (51.96%). In order to test Hypothesis 3, four groups were formed from the Male-Female and Thinking-Feeling combination, resulting in Thinking Males (n = 62, 30.39%), Feeling Males (n = 39, 19.12%), Thinking Females (n = 36, 17.65%), and Feeling Females (n = 67, 32.84%).
Measures
The research questionnaire included the Person-Activity Fit Diagnostic survey (Lyubomirsky, 2008) that measures happiness activities fitting one’s interests, values, and needs. The survey asks the degree to which the participant considers each of 12 happiness activities and to reflect on what it would be like to do it every week for an extended period of time. They are then asked to rate each activity on a 1 (not at all) to 7 (very much) scale regarding five reasons why they may engage in each activity. Thus, participants were asked to rate why they might keep doing this activity in terms of each of the following reasons:
-
Natural: I’ll keep doing this activity because it will feel “natural” to me and I’ll be able to stick with it.
-
Enjoy: I’ll keep doing this activity because I will enjoy doing it; I’ll find it to be interesting and challenging.
-
Value: I’ll keep doing this activity because I will value and identify with doing it; I’ll do it freely even when it’s not enjoyable.
-
Guilty: I’ll keep doing this activity because I would feel ashamed, guilty, or anxious if I didn’t do it; I’ll force myself.
-
Situation: I’ll keep doing this activity because somebody else will want me to or because my situation will force me to.
In addition to responding to the MBTI Form M written survey, the students participated in a “true fit” workshop. The purpose of a “true fit” workshop is for students to verify their personality type preferences and learn more about their types. For those people who have only a slight preference for Thinking versus Feeling in their written scores, they are able to consider which classification would be their best fit and re-classify themselves if they feel it is necessary.
Data Analyses
Table 1 presents the sample means and standard deviations for each of the 12 happiness activities and associated reasons for their choice (Lyubomirsky, 2008). The mean values show some differences in reasons for choosing happiness activities. In particular, choosing activities due to guilt or due to the situation were rated lower than choosing activities due to being natural, enjoyed or valued by the person.
Table 1
Sample Means of Five Possible Reasons for Choosing 12 Happiness Activities (1 = not at all, 7 = very much)
| Natural | Enjoy | Value | Guilty | Situation | ||
|---|---|---|---|---|---|---|
| 1. | Expressing Gratitude | 5.26 | 5.12 | 5.34 | 2.92 | 2.94 |
| 2. | Cultivating Optimism | 3.86 | 4.13 | 4.22 | 2.06 | 3.06 |
| 3. | Avoiding Overthinking and Social Comparison | 4.09 | 3.87 | 4.38 | 2.88 | 3.51 |
| 4. | Practising Acts of Kindness | 5.96 | 6.15 | 6.04 | 3.09 | 3.37 |
| 5. | Nurturing Social Relationships | 5.03 | 5.20 | 5.38 | 3.13 | 3.36 |
| 6. | Developing Strategies for Coping | 4.71 | 3.85 | 4.81 | 2.82 | 3.92 |
| 7. | Learning to Forgive | 3.82 | 3.56 | 4.54 | 3.22 | 3.57 |
| 8. | Increasing Flow Experiences | 5.20 | 5.72 | 5.25 | 2.41 | 3.28 |
| 9. | Savouring Life's Joys | 5.53 | 5.84 | 5.60 | 2.43 | 2.94 |
| 10. | Committing to Your Goals | 5.34 | 5.46 | 5.89 | 3.36 | 3.79 |
| 11. | Practising Religion and Spirituality | 3.83 | 4.22 | 4.75 | 3.59 | 3.33 |
| 12. | Taking Care of Your Body | 5.39 | 5.66 | 6.03 | 3.95 | 3.54 |
Hypothesis 1 proposed that reasons for choosing happiness activities would vary significantly by gender, and significant t-values lend support for five of the 12 activities (practising kindness, nurturing relationships, learning to forgive, savouring joys, and committing to goals). As shown in Table 2, female participants reported significantly more use of these five activities due to ‘natural feeling’ (practising kindness, t = -2.48, p < .01; nurturing relationships, t = -2.86, p < .01; savouring joys, t = -3.17, p < .002; and committing to goals, t = -2.09, p < .04), ‘enjoyment’ (practising kindness, t = -2.72, p < .01; nurturing relationships, t = -2.24, p < .03; and savouring joys, t = -2.62, p < .01), and ‘situation’ (learning to forgive, t = -2.03, p < .04).
Table 2
Significant Differences in Strategy Choice by Gender, Personality, and the Gender-Personality Interaction
| t-test Results
|
F-test Results
|
||||||
|---|---|---|---|---|---|---|---|
| Male (M) vs. Female (F)
|
Feeling (F) vs. Thinking (T)
|
Combined Types Differences
|
|||||
| Reason | Direction | Reason | Direction | Reason | Direction | ||
| 1. | Expressing Gratitude | Enjoy | F > T** | Enjoy | FF > MT* | ||
| 2. | Cultivating Optimism | Enjoy | F > T* | ||||
| 3. | Practising Acts of Kindness | Natural | F > M** | Natural | F > T** | Natural | FF > MT* |
| Enjoy | F > M** | Enjoy | F > T** | Enjoy | FF > MT** | ||
| 4. | Nurturing Social Relationships | Natural | F > M** | Natural | F > T** | Natural | FF > MT* |
| Enjoy | F > M* | Enjoy | F > T** | Enjoy | FF > MT* | ||
| Value | F > T** | ||||||
| 5. | Learning to Forgive | Situation | F > M* | Natural | F > T* | ||
| Value | F > T** | ||||||
| Guilty | F > T* | ||||||
| 6. | Savouring Life's Joys | Natural | F > M** | Natural | F > T** | Natural | FF > MT** |
| Enjoy | F > M** | Enjoy | F > T* | Enjoy | FF > MT* | ||
| Guilty | MF > FF* | ||||||
| 7. | Committing to Your Goals | Natural | F > M* | Guilty | T > F* | ||
| 8. | Practising Religion and Spirituality | Natural | F > T* | Natural | MF > MT* | ||
| Enjoy | F > T* | Enjoy | MF > MT* | ||||
| Value | F > T* | Value | MF > MT* | ||||
| 9. | Taking Care of Your Body | Situation | FT > MT* | ||||
Note. FF = Female Feelers; MF = Male Feelers; FT = Female Thinkers; MT = Male Thinkers.
*p < .05. **p < .01.
Hypothesis 2 proposed that reasons for choosing happiness activities would vary significantly by personality, operationalised as Thinkers versus Feelers. The hypothesis was supported by a significant t-value for eight of the 12 activities. As shown in Table 2, Feeler participants reported significantly more use of seven of these eight activities due to ‘enjoyment’ (expressing gratitude, t = -3.23, p < .001; cultivating optimism, t = -2.31, p < .02; practising kindness, t = -2.66, p < .01; nurturing relationships, t = -2.69, p < .01; savouring joy, t = -2.32, p < .02; and practising religion/spirituality, t = -2.27, p < .02), ‘natural feeling’ (practising kindness, t = -2.58, p < .01; nurturing relationships, t = -3.27, p < .01; learning to forgive, t = -2.05, p < .04; savouring joy, t = -3.14, p < .002; practising religion/spirituality, t = -2.36, p < .02), ‘value’ (nurturing relationships, t = -2.59, p < .01; learning to forgive, t = -2.80, p < .01; practising religion/spirituality, t = -2.09, p < .04), and ‘guilty feeling’ (learning to forgive, t = -2.21, p < .03).In addition, Thinker participants chose to commit to goals due to ‘guilty feeling’ significantly more than Feeler participants.
Hypothesis 3 proposed that strategy choice should differ significantly in four types combining gender and personality differences. Specifically, Females with the “Feeling” MBTI classification and Males with the “Thinking” classification were predicted to exhibit the largest number of differences in strategy choices, versus Females with the “Thinking” MBTI classification and Males with the “Feeling” MBTI classification. ANOVA verified significant differences exist between the FF, FT, MF, and MT types for five activities. Table 2 shows the four types differ significantly in how they express gratitude, practise acts of kindness, nurture social relationships, savour life’s joys, and practise religion/spirituality. As hypothesised, the significant differences were primarily between Female Feelers (FF) versus Male Thinkers (MT).
Discussion and Conclusions
The results of this study have important implications for understanding the relationship between gender and individual behaviour. The outcomes apply specifically to happiness/subjective well-being research, indicating that the reasons individuals choose some happiness strategies depend not only on gender and personality alone, but also on an interaction between them. Researchers who study happiness should consider gender as an important variable in theoretical models and as potentially interacting with personality. More broadly, the results of this study may support the view that personality is an underlying ‘deep-level’ diversity that provides a more-defined explanation for surface-level gender differences. The Thinking-Feeling dimension of the MBTI was used in this study, but other researchers should consider what additional deep-level dimensions might underlie such gender differences as individual values or attitudes toward work and relationships.
The results in this study suggest that personality gives greater dimensionality to the bipolar notion of female versus male. Although gender is often studied as an explanatory variable for behaviour, attitudes, or values, researchers rarely take issue with its lack of depth or its range restriction. Results here specifically show that the male-female dichotomy is a poor proxy for the underlying personality differences that explain choice of strategies for happiness. Gender may be a significant predictor of behaviour, as in this study, but its limitations obscure complexities that may be essential for full-specified models of human behaviour.
Interestingly, Female Feelers repeatedly reported that they pursued happiness strategies because they ‘felt natural’ and were ‘enjoyable,’ significantly more so than Male Thinkers reported. Male Feelers rated that they ‘valued’ strategies significantly more than Male Thinkers did. The addition of the Thinking-Feeling dimension, therefore, reveals these stereotypical differences are not female versus male, but they apply to subsets of each gender. Because most ‘feelers’ are female and most ‘thinkers’ are male, variation in strategy choice would have been attributed to gender if personality was not measured. The Male Feeler versus Male Thinker differences were not observed until personality was added to the analysis and would have been obscured in a study that only measured gender. As discussed above, a gender-only approach to research may wash out the underlying ‘deep-level’ explanation for behaviour.
More research questions arise as to why individuals pursue certain types of happiness strategies. Answering these questions could lead managers or counsellors to provide a broader set of strategies for pursuing goals that will appeal to more people. In practice, these preferred strategies may lead to more successful goal attainment. What, ‘feels natural’ and ‘enjoyable’ to Female Feelers? The strategies appear to be typically categorized as female, including expressing gratitude, practising acts of kindness, nurturing social relationships, and savouring life’s joys. Past researchers concluded that individuals may develop deeply-held values for certain behaviours and traits if they adopt stereotypical values and then respond to psychological tests in a way that is consistent with these biases (Eagly, 1987; Mazzella & Feingold, 1994; Saad & Sackett, 2002). In fact, happiness may be perceived as a feminine concept altogether. Alternately, the individual may define happiness and definitions can vary depending on one’s personality. Another explanation for stereotypical gender differences may be that the subsets that actually adopt societal norms also dominate within the stereotyped gender. The whole male group or female group takes on the broad characteristics of the majority. In fact, research shows that gender has particular problems with prediction due to the differences between gender and gender-role orientation (Choi, 2004).
Results of this study also provide information for gender in management practice. Clearly, the reasons for pursuing various happiness strategies are not solely dependent on gender, but also on personality. An effective manager would not assume, for example, that a female employee is regularly kind to coworkers because it is a natural behaviour, when she may be acting solely based on specific situational factors. An effective manager would also not assume that a female employee would express gratitude more often because she finds it more enjoyable than male employees. The lack of awareness of differences represented by Female Thinkers (FT) and Male Feelers (MF) also may lead to false attributions.
Female Thinkers (FT) tend to be more detached emotionally from a decision situation and prefer logical analysis. The FT may be compared to Female Feelers (FF) as being less personable and be negatively judged. Or, Female Thinkers (FT) may be more positively judged in an environment that favours logic and practicality. Similarly, Male Feelers (MF) who are more attuned to the harmony and balance needed in a decision situation may be compared to the perceived male standard of Male Thinkers (MT) and subsequently judged as lacking logic or knowledge. Alternately, Male Feelers (MF) may excel in an environment heavily reliant on interpersonal relationships and values.
In conclusion, this study indicates that the relationship between personality and gender provides a more complete explanation for choice of happiness strategies. Specifically, reasons for choosing particular strategies were significantly different for a subgroup of women versus a subgroup of men. These would have appeared as simple gender differences if personality was not measured.
Future researchers should investigate other measures of gender orientation (Bem, 1981). Future researchers should determine how gender orientation interacts with personality to predict behaviour. Future researchers may be interested in why differences occurred in the reasons for choosing strategies. For example, Female Feelers (FF) rated ‘enjoyment’ significantly higher than Male Thinkers (MT) as a reason for four strategies (expressing gratitude, practising kindness, nurturing relationships, and savouring joy). FF also rated ‘a natural feeling’ significantly higher than MT as a reason for using three strategies (practising kindness, nurturing relationships, and savouring joy). In the workplace, researchers could explore in the future the effects of gender and personality interactions on job design, morale, training, career counselling, benefits options, or reward systems.
Limitations of this research include the use of a self-report measure of strategy choice versus actual choice behaviour. Participants were required to rely on their memories of past choices, which memories may be flawed. Another limitation may be present in the cross-sectional nature of the research, since a longitudinal approach could be better fitted to assess cause-effect relations. Despite these limitations, however, this study may provide one step forward in providing deeper-level explanations for gender differences in behaviour.