I took the long-form MMPI-2 today using the procedure given in , and scored it for both female and male using the same answers. My goal is to see what differences emerged between the sexes in the outcome. My answers and the numerical results are attached to this blog post.
How I Addressed the Questions Related to Gender Identity
There are two questions related to gender identity in the long-form MMPI-2:
#62 – I have often wished I were a girl. (Or if you are a girl) I have never been sorry that I am a girl.
#371 – I have often wished I were a member of the opposite sex.
I answered #62 true because I am happy being a woman. I also answered #371 true because I considered the whole arc of my life when answering the question—considered that I’ve spent most of my life wanted to be a woman.
I’m sure this messes with the “Gender Role” and “Masculinity-Femininity” scales somewhat.
VRIN scores for both sexes were fine . TRIN raw scores were one greater than the recommended cutoff, suggesting a retake of the test is warranted. However, I made every effort to take the test honestly and consistently, so I first examined the questions that impacted my TRIN score:
|Paired Question #1||Anwser to #1||Paired Question #2||Answer to #2|
|I have very few quarrels with members of my family.||True||My parents and family find more fault with me than they should.||True|
|I believe that my home life is as pleasant as that of most people I know.||True||There is very little love and companionship in my family as compared to other homes.||True|
|I am not happy with myself the way I am.||True||If I could live my life over again, I would not change much.||True|
|I worry a great deal over money.||True||I am satisfied with the amount of money I make.||True|
As I see no real contradiction in my answers to above stated questions, I decided to proceed with these results rather than retake the test.
There are two gender stereotype scales: “Gender Feminine” and “Gender Masculine”. They are rather cultural. Higher scores for Gender Masculine correspond to:
“Traditional attributes of masculine strength, such as self-confidence, forthrightness, goal persistence, and freedom from fears, worries, self-consciousness, and social inhibition.” 
By contrast, higher scores for Gender Feminine correspond to:
“Traditional feminine attributes of social circumspection, agreeableness, trust, loyalty, and the avoidance of conflict and impropriety. 
Here are my scores for these scales. For some reason the scoring calculation could not define a Gender Feminine score for my female-scored test:
What definitely sticks out is my high femininity score when scored as a male.
Scored as a female, my masculinity is modest and below average. Scored as a male it is also modest, but this time above average.
The “Masculinity-Femininity” scale measures how likely one is to reject traditional gender roles . My scores follow:
While not to a great extreme, I reject traditional gender roles when scored as a female or as a male.
Scales Scoring More Than One Standard Deviation In Difference Between the Sexes
Two scales, “Antisocial Behavior” and “Disconstraint” produced a score difference greater than one standard deviation between the female and male tests for the same answers. The female side dominated in both:
Antisocial behavior measures rule-breaking and “acting out”, such as failure to conform . I’ve always been somewhat of an anarchist! But I suspect the score difference here reflects simply the fact that the women in the calibration set were more conservative than the men.
 Gives a nice description of disconstraint:
Persons high on this scale show (a) higher levels of physical risk- taking, (b) have a style characterized more by impulsivity than control, and (c) are less bound by traditional moral constraints. 
Yep, that is me! Especially after becoming a woman! Again, I suspect the score difference here reflects simply the fact that the women in the calibration set were more conservative than the men.
Clearly I’m trouble as a woman! But well-behaved women never make history.
I’m Equally Well-Adjusted When Scored Either Way
For each scale, I calculated the difference between the male score and the female score, and then plotted the distribution of the differences:
While the median leans slightly on the male side, the mean is indistinguishable from zero as per a t-test.
It is interesting that all the outliers are on the female side.
But I’m an outlier… and I walk on the feminine side!
Data and Analysis Code
- Nichols, David S., Essentials of MMPI-2 Assessment. John Wiley & Sons, Inc. 2001