Study A - Masculinity and Femininity

Possible rationale for a re-investigating this measure:

Arguably, sex roles have changed in the forty years since Bem devised and validated this questionnaire measure. You should be able to find sources that provide evidence of this. Such changes may have altered aspects of the reliability and validity of the masculinity and femininity scales.

Some general background/context for the ideas related to this study:

Psychology and the Individual material on personality dimensions; the joint RM2/Individual practical class (Week 3); and perhaps Psychology and Society material on self identify.

The data collection for this study:

In Week 1 of term ('time 1'), students in the three RM2 practical classes completed a pencil-and-paper version of the BSRI. This had 60 items, which were answered on a 7-point scale, and were presented in the same fixed order for all participants. There were 20 items each for the BSRI masculinity and femininity scales and the social desirability scale.

In Week 2 of term, students in the three RM2 practical classes completed an on-line questionnaire; Qualtrics software was used to present the questions. First, students were asked to indicate their sex and their study code for linking data. Then students completed questions for Study B. Then students completed the BSRI, with the 60 items presented in the same fixed order as the previous week. Finally, students completed two 20 -item personality scales (autonomy and agreeableness) from the Five Factor Personality Inventory (FFPI; Hendriks, Hofstee & De Raad, 1999); items alternated between those from the autonomy and agreeableness scales. All students encountered these questions in the same fixed order.

The data file for this study includes:

The sex of the participant 60 variables that record the responses for the 60 BRSI items at time 1 .

Recoded responses for 10 'low' social desirability items (e.g., 'moody') at time 1.

60 variables that record the responses for the 60 BRSI items at time 2 .

Recoded responses for 10 'low' social desirability items (e.g., 'moody') at time 2 .

40 variables that record the responses for the autonomy and agreeableness items.

Recoded responses for 10 'low' autonomy items (e.g., 'does as he/she is told') and for 10 'low' agreeableness items (e.g. 'gets angry easily')

Overall scores for BSRI masculinity and for BSRI femininity, at time 1 and time 2 (psychological androgyny difference scores for each time point can be calculated from these).

Overall FFPI agreeableness and FFPI autonomy scores.

Suggested data analysis for this study

1. Examine the internal consistency of the masculinity and femininity scales. If a multiple-item measure is internally consistent, the responses to the items should correlate with each other. Cronbach's alpha is the most common technique for measuring the internal consistency of a measure, and you can also use SPSS to identify those items that do not fit well with the other items from the same scale. For this (and other analyses) you can use the time 1 BSRI responses, or the time 2 BSRI responses, or both sets of responses when you look at this.
2. Examine the test-retest reliability of the masculinity, femininity and psychological androgyny (PA) scales. This involves examining how strongly scores obtained at time 1 correlate with scores obtained at time 2 for the same measure. A strong positive correlation in usually taken to indicate that a scale has good test-retest reliability.
3. Bem had very specific ideas about the nature of masculinity and femininity: these are separate ('independent', or 'unrelated') dimensions of personality, not opposite ends of a single dimension of personality as had previously been supposed. If true, the correlation between masculinity scores and femininity scores should be close to zero. (Note: it is not possible to 'prove' that a correlation is zero; e.g., a non-significant correlation does not prove that two variables are uncorrelated, rather it shows that we do not have good evidence that they are correlated. Therefore, be careful when writing this up.) If, counter to Bem's ideas, masculinity and femininity scores are strongly correlated, you could explore whether it would be better to have a single scale that had items from both scales than to have two separate scales.
4. Masculinity and femininity are psychological constructs that reflect dispositions or tendencies to behave in particular ways, and therefore are not the same as biological sex. Nonetheless, it would be an odd choice of name for these constructs if men were not, on average, more masculine (and less feminine) than women. You can use a \(t\) -test to examine whether males and females differ (on average) in their masculinity, femininity and psychological androgyny scores as would be expected.
5. The BSRI masculinity and femininity scales were designed to contain desirable traits (those especially desirable for a man, or for a woman, respectively). Therefore, we expect some positive correlation between social desirability scores and scores on the masculinity and femininity scales. However, the correlation should not be so strong as to suggest that masculinity and/or femininity are simply alternative names for social desirability. You can also examine, as Bem did, whether psychological androgyny scores correlate with social desirability.
6. Several BSRI items are similar to items that can be found on questionnaires that measure major dimensions of personality. For example, several BSRI femininity items seem to be similar to items used elsewhere to assess agreeableness, and several BSRI masculinity items seem to be similar to items used elsewhere assess to autonomy or independence. It is therefore important to check that Bem is not simply measuring what others call autonomy (and labelling it as 'masculinity') or what others call agreeableness (and labelling it as 'femininity'). If there is a very strong positive correlation between these measures, this could be argued to be the case. Such analyses (e & f) are considering the divergent validity of the BSRI (i.e., checking that they BSRI is measuring something different to other established measures).

Price: $49.99

Solution: The downloadable solution consists of 45 pages, 656 words.
Deliverable: Word Document