Dissertation Summary: Changes in the Coming Out Process

INTRODUCTION

 

Changes in the Coming Out Process Over Time, a brief summary

Andy Dunlap PhD, LCSW

1/15/2012

 

 

This research summary is designed to disseminate findings of an internet research study:  An Investigation into Changes in the Coming Out Process Over Time.  The study is complete and the findings are presented here in summary form.

 

 

Many people participated in this research either directly or by spreading the word about the survey.  Please spread the word about the findings to anyone you might have passed the survey on to.

 

 

I would like to thank everyone who supported this research project over the past several years.  This includes the 1131 participants in the study, the 115 LBGT Community Organizations that helped to distribute the internet survey, and everyone else who helped get the word out.

 

 

This dissertation project was completed on October 18th, 2011 with a successful defense hearing.  All revisions to the text were also completed on that date.  Special thanks to David Burton, Phebe Sessions and Marsha Kline Pruett for their support on this project.

 

 

The full text of the dissertation (in electronic form) can be found through interlibrary loan via the Smith College Libraries.

 

 

Finally, thank you to my partner, Scott Manning, for supporting me in this and all endeavors.

 

 

ABSTRACT

This mixed methods study examined changes in the coming out process for same-sex attracted people through three interrelated research questions.  An internet survey was used to recruit 1131 participants (ages 19 to 85) who self-identified as members of the lesbian, gay and bisexual community.  Participants provided demographic information, ages that they achieved significant milestones in the coming out process, responded to five open questions about their coming out experience, and completed the Lesbian, Gay, Bisexual Identity Scale (LGBIS) (Mohr & Fassinger 2000).

 

In research question one, significant differences were found between participants born before 1989 and all other participants on several demographic variables and scores on the LGBIS.  In general, 19 to 22 year olds in this study rated more difficulty in the coming out process than older cohorts on this measure.  Findings for research question two were a qualitative thematic analysis of a stratified random sample (n = 30) of the larger data set.  Major themes that emerged from this analysis included two paths to coming out and the importance of supportive relationships as facilitative of the coming out process.  The final research question involved analysis of variance of significant milestones in the coming out process between men and women sorted over five historic age cohorts.  Findings included significant differences between men and women.  Differences in the order of achieving milestones were observed between cohorts and genders.  Finally, that the coming out process has moved from a largely adult process to an adolescent one was confirmed and explored in these findings.

 

 

 

SAMPLE

This study utilized a convenience snowball sampling strategy that relied largely upon the LGBT community as a means for distributing the survey.  I contacted 115 LGBT community organizations across the country and asked for their help in distributing the survey.  I also sent my plea out to social work research networks and also asked my professional contacts to distribute the survey as well.  Between the dates of November 10th, 2010 and October 19th, 2011 1336 people visited the survey web page.  Of those, 955 (71.5%) completed each question of the survey.  For several of the more salient data points (see Research Question 3), 1131 people responded.

 

 

 

 

RESEARCH QUESTION ONE:  What is the coming out experience of today’s same-sex attracted youth?

In order to investigate this question I sorted my respondents into two groups:  those born in 1988 and earlier and those born after 1988.  Essentially, I compared all 18 to 22 year olds in the sample to all other members of the sample.

Age and Gender with Two Cohorts

Female

Male

Total

N (%)

M; SD

N (%)

M; SD

N (M; SD)

Youngest Group:  born after 1988

140 (65.1%)

20.52; 1.16

75 (34.9%)

20.89;  1.05

215 (20.65; 1.14)

Older Group:  born1988 and earlier

507 (55.5%)

40.66; 13.60

406 (44.5%)

47.32; 14.17

913 (43.64; 14.24)

Total

647 (57.4%)

36.24; 14.63

481 (42.6%)

43.22; 16.17

1128 (39.23; 15.69)

 

I found significant differences between the older group and the younger group on four demographic variables.

Independent T Tests of Variables Between Cohorts

 

N

M

SD

ta

Growing up, how socially liberal or how socially conservative was your household?b

younger cohort

209

4.17

1.87

3.39***

older cohorts

897

3.68

1.89

Growing up, how important was organized religion in your household?c

younger cohort

220

4.23

2.13

3.24***

older cohorts

895

4.75

2.04

(continued)

Independent T Tests of Variables Between Cohorts (continued)
At first, how supportive was your childhood primary care giver of your sexual orientation?d

younger cohort

187

3.90

2.03

2.91**

older cohorts

796

3.43

2.01

Now, how supportive is your childhood primary care giver of your sexual orientation?d

younger cohort

186

5.00

1.92

2.39**

older cohorts

706

5.36

1.80

a results presented in order of significanceb on a seven point scale with conservative = 1 and liberal = 7c on a seven point scale with not important = 1 and very important = 7

d on a seven point scale with not supportive = 1 and very supportive = 7

** significant at the .05 level

*** significant at the .001 level

 

The younger group in this sample grew up in households that were significantly more socially liberal and in which organized religion was significantly less important.  In my sample, the younger group reported significantly higher levels of initial support from primary care giver regarding their sexual orientation.  I was also surprised to find that the younger group in my sample reported lower levels of current support from their primary care giver than the older group.  These findings suggest that the youngest group in my sample moved through the coming out process in a more supportive environment.

 

That the older cohort reported higher levels of support from the primary care giver may be interpreted in a number of ways.  Since the amount of initial support for the older group was lower, perhaps time has allowed for these parents to become more supportive, describing an upward trend in familial support.  Primary caregivers of younger participants in this study may not have had enough time to change the amount of support that they provide for their children.

 

The primary data that I used in this analysis was average scores on the subscales of the Lesbian, Gay, Bisexual Identity Scale (LGBIS) (Mohr & Fassinger 2000).  The six subscales are designed to measure different levels of difficulty in the coming out process.

 

  Independent T Tests of LGBIS Scores Between Youngest Cohort and older Cohorts

 

N

M

SD

t

Subscale1:Internalized Homo/Bi Negativity

younger cohort

144

2.02

1.27

2.17a**

older cohorts

732

1.77

1.02

Subscale2:Need for Privacy

younger cohort

146

4.06

1.24

4.20***

older cohorts

731

3.60

1.20

Subscale3:Need for Acceptance

younger cohort

145

3.04

1.23

4.64a***

older cohorts

736

2.53

1.12

Subscale4:Identity Confusion

younger cohort

147

2.17

1.44

5.25a***

older cohorts

745

1.52

0.97

Subscale5:Difficult Process

younger cohort

145

3.83

1.41

2.13a**

older cohorts

731

3.55

1.53

Subscale6:Superiority [over heterosexuals]

younger cohort

148

2.13

1.15

2.01a**

older cohorts

748

2.34

1.32

Composite Scale:Negative Identity

younger cohort

139

3.27

0.98

4.79***

older cohorts

705

2.86

0.89

aequal variances not assumed** p < .05***p < .001

 

There were significant differences between the two groups in my sample on all measures of the coming out process.  In all but one, Subscale 6, Superiority [over heterosexuals], the younger group reported higher scores (greater difficulties) than the older group.   Analysis of this data suggests that the 18 to 22 year olds in this sample came out in more accepting families, but struggled more than their predecessors.

 

But what about the confounding factor of maturity?  Was it possible that older members of my samples reported fewer difficulties in the coming out process because they had simply had more years to work through this process?  Could age account for lower scores on the LGBIS?  In order to find out I performed a further analysis of these variables [not included in the actual dissertation].  This took the form of investigating correlations between age within the entire sample and scores on the LGBIS.  If scores were negatively correlated (higher age = lower scores) then age was perhaps a better explanation for the significant differences in these scores rather than more struggles for the younger group compared to the older group.

Correlations between age and LGBIS scores

Significance

Subscale1:Internalized Homo/Bi Negativity

r(868) = -.042, no significant correlation

Subscale2:Need for Privacy

r(869) = -.096, p (2 tailed) < .05

Subscale3:Need for Acceptance

r(872) = -.228, p (2 tailed) < .001

Subscale4:Identity Confusion

r(884) = -.298, p (2 tailed) < .001

Subscale5:Difficult Process

r(868) = -.034, no significant correlation

Subscale6:Superiority [over heterosexuals]

r(888) = -.001, no significant correlation

Composite Scale:Negative Identity

r(836) = -.119, p (2 tailed) < .001

 

For several of these LGBIS scores there was a significant correlation between age and the score.  Need for Privacy, Need for Acceptance, Identity Confusion and Negative Identity were all negatively correlated with age.  Scores tended to decrease as respondents’ age increased.  This suggested that age might account for differences between the two groups on these variables.  It should be noted that theses correlations, while significant, were all weak (r < .5).

 

Interestingly, three of these subscales were not significantly correlated with age.  Internalized Bi/Homophobia, Difficult Process, and Superiority [over heterosexuals] had no correlation with age.  This suggests that age does not account for differences in these scores.

What is the coming out experience of today’s same-sex attracted youth?  These findings suggest that 18 to 22 year participants in this study are coming out in families that are more accepting, but that they are having a more difficult process and experiencing a greater degree of internalized bi/homophobia.

 

What might account for this seeming contradiction?  See Research Question Three for a possible answer.

 

 

 

 

RESEARCH QUESTION TWO:  What are the personal theories that same-sex attracted individuals have about the coming out process?

 

In this portion of the study, I asked participants to respond to the following five questions in an open format.

  1. How do you label your sexual orientation and how did you come to that conclusion? What helped/helps you figure out your sexual orientation?  (people, supports, experiences).
  2. Did you have (or do you have) any personal difficulties related to understanding and becoming comfortable with your sexual orientation?  (for example: depression, lost friends, drinking too much, etc.) If you experienced these or other problems, please describe them.
  3. If you’ve had any personal difficulties related to understanding and becoming comfortable with your sexual orientation, who or what helped you get through these difficulties?
  4. How has the process of understanding and becoming comfortable with your sexual orientation added to your life?  What, if any, benefits have there been?  (for example: shaping who you are, impact on relationships, etc.).
  5. Do you think the process of understanding and becoming comfortable with one’s sexual orientation is different now than it used to be?  Why do you think that is the case?

 

Because there were a large number of responses to each of these questions (960; 956; 870; 882; and 883 respectively) I decided to do a thematic analysis of a stratified random subset of the data set.  I began by applying my five cohort sampling plan.

Cohorts
Years born Current ages Historical frame of reference**
Cohort A Before 1951 60 and older Pre-Stonewall Generation
Cohort B 1951 – 1962 59 to 49 Stonewall Generation
Cohort C 1963 – 1969 48 to 42 AIDS Crisis Generation
Cohort D 1970 – 1988 41 to 23 Post AIDS Crisis and Millennial Generation
Cohort E After 1988 22 – 18*** Youngest Cohort
***respondents under the age of 18 were screened out

 

and generated a stratified random sample of 30 respondents for thematic analysis.

Ages of Qualitative Subsample

Women

Men

Total

N; M (SD)

N; M (SD)

N; M (SD)

Cohort A

3; 62.7 (1.53)

3; 70.0 (7.81)

6; 66.3 (6.44)

Cohort B

3; 52.0 (2.65)

3; 59.7 (1.53)

6; 51.3 (2.07)

Cohort C

3; 44.7 (2.89)

3; 44.7 (1.53)

6; 44.7 (2.07)

Cohort D

3; 30.0 (4.58)

3; 28.3 (4.73)

6; 29.2 (4.26)

Cohort E

3; 20.0 (1.00)

3; 21.0 (1.00)

6; 20.5 (1.05)

Total

15; 41.9 (15.96)

15; 42.9 (18.22)

30

 

In general the responses created a narrative arc across all five questions.  As the questions themselves have a progression, respondents tended to follow the outline of this inquiry.  An important tentative theme that emerged from this small thematic analysis was the notion of two paths of coming out:  People who figured it out in their heads, and those who figured it out in their hearts.

 

Those who figured it out in their heads described coming out as a process of studying, reading, learning about the community and coming to the conclusion that their internal experience matched what they learned about being an LGB person.  Those who figured it out in their hearts shared narratives of falling in love or having sex and then “just knowing.”

 

Labels are restrictive:  Another important tentative theme was that of the problematic nature of labels.  Some respondents felt that the processes of labeling others and self-labeling where restrictive.  At the same time, some of these respondents also acknowledged the necessary evil of labels for many people in the service of coming to understand their sexual orientation.

 

The types of personal difficulties that these thirty respondents had grouped into themes that were consistent with the literature on the topic:  mental health problems; identity confusion; cultural violence; and loss of connectionCultural violence included any physical, verbal or psychological attack towards the respondent because of sexual orientation.  Loss of connection included any shunning or disconnection from family or significant others because of sexual orientation.

 

Relationships; professional help; education; and entering community were the four themes that emerged from this subsample describing factors that facilitated the coming out process.  This was also consistent with the literature on the topic.

 

The benefits of working through a coming out process grouped in to two themes:  enhanced personal development and greater empathy.  Enhanced personal development included many aspects of greater self-awareness and the development of strengths.  Empathy included not only a greater understanding of the experience of others, but also an appreciation of diversity that might not otherwise have been experienced.

 

These thirty respondents provided a somewhat mixed response when asked if they felt the coming out process was easier than it used to be.  Of the twenty two who responded to the question, sixteen felt that it was now easier.  Eight were unsure.  A few noted that for any given individual, coming out can still be a lonely and isolating experience.  Many noted that the environment seemed to be more facilitated for coming out (citing the internet specifically), but that there still seems to be significant struggle (citing a seeming increase in LGB youth suicides).

 

 

 

RESEARCH QUESTION THREE:  How has the coming out experience changed over time?

 

For this final research question, I again used the five cohort model to sort my respondents.  This time, however, I used the entire data set in order to try to gain a quantitative understanding of differences in the coming out process.

Cohorts
Years born Current ages Historical frame of reference**
Cohort A Before 1951 60 and older Pre-Stonewall Generation
Cohort B 1951 – 1962 59 to 49 Stonewall Generation
Cohort C 1963 – 1969 48 to 42 AIDS Crisis Generation
Cohort D 1970 – 1988 41 to 23 Post AIDS Crisis and Millennial Generation
Cohort E After 1988 22 – 18*** Youngest Cohort
***respondents under the age of 18 were screened out
 

 

Age and Gender with Five Cohorts

Female

Male

Total

N (%)

M; SD

N (%)

M; SD

N (M; SD) %

Cohort A

47 (7.2%**)

66.8; 5.4

80 (16.6%)

66.6;  5.1

127 (66.7; 5.2) 11.2%

Cohort B

113 (17.4%**)

53.8; 3.6

124 (25.8%)

54.3; 3.4

237 (54.1; 3.5) 21.0%

Cohort C

68 (10.5%)

45.3; 2.0

51 (10.6%)

45.4; 2.1

119 (45.3; 2.1) 10.5%

Cohort D

282 (43.5%)

30.2; 5.1

151 (31.4%**)

31.8; 5.7

433 (30.8; 5.3) 38.3%

Cohort E

140 (21.5%)

20.5; 1.2

75 (15.6%**)

20.9; 1.1

215 (20.6; 1.1) 19.0%

Total

650 (100%)

36.3; 14.7

481 (100%)

43.2; 16.2

1131 (39.3; 15.7)

 

The range of ages was between 19 years of age and 85 years of age.  The goal of data collection was to achieve a representative sample of fifty men and fifty women over five historic groupings.  Only one cohort did not reach the goal of 50 subjects, women born before 1951 (n = 47).

 

In this sample, fewer women over the age of 49 are represented than would be expected in a random distribution and fewer men under the age of 41 than would be expected are represented in this sample.  Additionally, as a whole, Cohorts B and D are overrepresented within the sample as a percentage of responses.  That being said, the type of analysis I conducted concerned itself with the average ages of each of these cohorts and does not require equally sized groups for analysis.

 

The main analysis for this question was a series of ten ANCOVA analyses of important milestones in the coming out process.

 

I analyzed several likely covariates for inclusion in this analysis.  Please see the dissertation proper if you would like more information on this process.  I identified the following factors as significant covariates in at least one of the examined milestones.

  • How socially liberal or conservative was your household growing up?
  • How important was religion in your household growing up?
  • Now, how supportive is your primary caregiver of your sexual orientation?
  • Race
  • Internalized Homo/Biphobia (as measured by the LGBIS)
  • Need for Privacy (as measured by the LGBIS)
  • Need for Acceptance (as measured by the LGBIS)
  • Identity Confusion (as measured by the LGBIS)
  • Difficult Process (as measured by the LGBIS)
  • Superiority [over heterosexuals] (as measured by the LGBIS)

 

Full details of the ANCOVA analysis are available in the dissertation.  Differences in average ages of reaching each milestone are presented below in graphic form.

 

 

 

 

 

 

 

 

 

 

 

I found significant overall differences in all 10 ANCOVA that I performed (p < .001).  Additionally the effect size was at least medium (partial eta squared > .09) to strong (partial eta squared > .25) for all analyses.  The partial eta squared ranged from .150 to .348 for various milestones.  This indicated that the interaction between cohort membership and gender accounted for 15% to 34.8 % of the overall variance for each milestone respectively.

 

I performed further post hoc analysis, using the Bonferroni test in order to determine which groupings were significantly different from each other. Fifty Two percent of the 450 pairwise comparisons were significant at least the .05 level.  Please see the dissertation for the details of these analyses.

 

Generally speaking, there were two significant trends that emerged from these analyses.  The first was a general decrease over time in average age of achieving significant milestones in the coming out process.  The second was a general narrowing of a gender gap between men and women in the average ages that milestones were achieved.

 

All milestones were achieved at younger ages by younger cohorts and consequently milestones were achieved in less time.  See the dissertation for a complete discussion of this trend.

 

 

 

DISCUSSION

The coming out experience for the youngest cohort in this study seems to be a bit of a mixed bag.  The youngest group reports a more supportive environment for coming out, while also reporting a more difficult process and a greater degree of internalized homo/biphobia compared to the older grouping.

 

People in this sample are reaching coming out milestones at younger ages and differences between men and women are disappearing.  As the age at which important coming out milestones are reached has decreased, the quality of the coming out experience has moved from more of an adult activity to an adolescent process.

 

The speed at which people in this sample moved through these milestones has also increased.  The youngest women in this sample moved through these milestones in an average of 5.2 years compared to the oldest women who spent 12.1 years during this process.  The difference between the oldest and youngest group of men was even greater at 22.4 years and 6.8 years respectively.

 

Perhaps a more facilitative environment has contributed to a decrease in the average age of coming out.  While this is can be seen as a positive development, it creates a perverse problem for young people who are coming out.  Traditionally, supportive resources for same-sex attracted youth have been targeted at young adults rather than teens (under 18).  Fewer supports seem to be available for struggling adolescents than for struggling young adults.

 

Today’s thirteen year old may be struggling with the same issues as yesterday’s twenty year old, but with fewer resources and a less solid sense of self.  This may account for an increase in difficulty in the coming out process while the age of achieving significant milestones has decreased.  Please see the dissertation for a complete discussion of these findings.

FURTHER RESEARCH AND MORE INFORMATION

The role of race/ethnicity as a facilitative/inhibitive factor in coming out is touched on in this dissertation, but is an area for further exploration within this data set.

 

The coming out experience of transgender people has not been explored in this research.  A sample of 59 transgender folk was collected but not included in this analysis.  I chose to segment this group because I believe that the transgender coming out experience is intertwined with the gender emergence process and would not have been adequately explored under the current research design.  This is an area for further research in this data set.

 

Labels that study participants use to describe their sexual orientation were explored within the dissertation, but not presented in this summary.

 

Roughly 900 narrative responses have not yet been evaluated within this data set and are only explored via the thematic analysis of a subsample (n=30).  This is an important area for further investigation within this sample.

 

Responses to the survey included many international participants.  Cross-cultural evaluation of these research questions is an important and interesting area for further exploration.

 

Quantitative differences between those that figured it out in their hearts and those that figured it out in their heads might lead to a fuller understanding of the individual and common aspects of the process of coming to understand and accept same-sex attraction as part of one’s identity.