Homeschoolers Anonymous - WordPress.com

2

Homeschool Alumni Reaching Out presents

A Complex Picture: Results of a 2014 Survey of Adult Alumni of the Modern Christian Homeschool Movement Installment 1: Descriptive Summary December 2, 2014 Data analysis provided by: Coalition for Responsible Home Education 661 Washington Street #563 Canton, MA 02021 (617) 765‐7096 [email protected]

3 About the Survey In 2014, Homeschool Alumni Reaching Out (HARO), the parent organization of Homeschoolers Anonymous (HA), conducted a survey of adult alumni of the modern Christian homeschool movement in consultation with the Coalition for Responsible Home Education (CRHE). The purpose of this survey was to investigate the life experiences of Christian homeschool alumni by collecting information that past surveys of homeschool alumni had not. The data collected will be used to advocate for the interests of current and future homeschooled children. The survey, written by R.L. Stollar, was developed over a span of 9 months. Work on the survey began on November 24, 2013 and it was opened to the public on August 18, 2014. A set of approximately 90 initial questions were first created. These questions were then tested, modified, and re‐tested repeatedly over a span of 6 months to create the survey questions that were on the final version. The questions were specifically run by a diverse group of people, including Christians and non‐Christians, conservatives, moderates, and liberals, homeschoolers and unschoolers, and so forth. The final version of the survey featured questions on demographics, academic school experiences, non‐academic school experiences, food and health, religion, present and future personal life plans, sexuality, mental health, and abuse. The introductory page of the survey is given in Appendix A. The survey, conducted online through SurveyMonkey, was estimated to take respondents 30 minutes to complete. It was first promoted through the homeschool abuse survivor community, from which it spread across the country through online social networks (primarily Facebook). Survey respondents were required to affirm that they were 18 years old or older, had been homeschooled for at least 7 years, were homeschooled in an environment which was classifiable as Christian (including Christian‐influenced new religious movements), and were taking the survey through completion for the first time. A total of 6,249 people started the survey; 3,702 respondents completed the survey before it closed on September 15, 2014. Only the completed responses were recorded and analyzed. Strengths of the survey over existing work The existing research on homeschooling outcomes is extremely limited. As such, this survey provides new insights into homeschool alumni’s experiences of sexuality, mental illness, and child abuse, topics about which there was only anecdotal data prior to this survey. By relying primarily on social networks of friends to recruit respondents, the survey’s respondents were more age‐, education‐, and racially diverse than Ray’s (2003) similar study. Unlike the similar Cardus study (2011), this study did not seek to compare homeschool alumni to graduates of other types of education—rather, it sought to determine how best to advocate for homeschooled children’s interests, in keeping with HARO’s mission to renew and transform homeschooling from within. As the survey was written and designed by homeschool alumni, it also provides a necessary perspective in the study of homeschooling—that of the homeschooled child. Homeschooled children have an interest in receiving care and protection from the negative events which plague every community—child abuse, mental illness, etc. As such, the underlying hypothesis of the survey was that these events may be occurring in the homeschooling community, contrary to claims made by some homeschooling advocates. It also sought to determine the prevalence of these negative events. Other than these sources of bias, there was no agenda in terms of proving correlations between variables.

4 Challenges in analyzing the data Despite the survey’s strengths, there are a number of challenges to data analysis which indicate that the study should not be used to “prove” any claims about homeschooling. 1) Non‐random sample A common problem with researching homeschoolers is that there is no central register from which to draw a random sample. Because of this, the respondent pool—recruited through friend networks—may be over‐ representative of certain communities. For instance, the survey originated in the homeschool abuse survivor community; this may lead to an over‐representation of homeschool alumni who reported child abuse. There was also a higher than expected response from asexual homeschool alumni and those in California and Texas. Three times as many women took the survey as men; it is unclear why this might be. Furthermore, as the survey was online, there may have been an under‐representation of homeschool alumni who lack a regular internet connection—those who lack home computers or regular employment, for example. 2) Non‐representative sample In addition to being non‐random, the sample of respondents is also not representative in terms of the general homeschool population. For instance, NCES data shows that around 68% of homeschoolers are white; by contrast, our sample was 85% white. The sample was also non‐representative of all homeschoolers in that only those who were homeschooled for 7 years or more were allowed to participate, and homeschool alumni raised in non‐Christian environments were not represented at all. 3) Technical difficulties A few respondents reported technical difficulties with the survey which may reveal additional sources of bias. For instance, the question asking for birth year was missing the answer ‘1988’ for the first several hours of the survey; as this was the most common birth year, some respondents may have been forced to give the wrong answer to this question. In addition, some respondents reported that the answers to their survey were already partially filled in—it is not clear why this happened, but it may have biased these respondents in some way. Because of these issues, this survey should not be used to make any generalized statements about homeschoolers as a whole; the only people it can be used to make claims about are the 3,702 people who took the survey. That said, many of the trends observed—particularly those which were less likely to be triggered by selection bias—may be suggestive of larger trends in the population and may indicate directions for future research. Thus the survey yields valuable insights into homeschool alumni’s experiences even if it cannot be used to prove any claims about them.

5 0. Introduction to Installment 1: Descriptive Summary In this installment, we will give a general overview of the major sections in the survey, showing basic descriptive statistics on the questions deemed to be most interesting to the homeschool alumni who took the survey. There will be no comparisons of variables or statistical tests in this installment. The major sections here include demographics, sexuality, family & work, academics, non‐academics, religion, mental health, and abuse. 1. Demographics This section includes a description of respondents’ age, gender, race, socio‐economic status, geographic location, and family structure. Age The mean and median of respondents’ birth years were both 1988, meaning that at the time of the survey our respondents’ mean age was 26. See Figure 1 below. Note that, for the first several hours of the survey, 1988 as a birth year was missing from the survey; the slight dip observed on that year is artificial and 1988 would most likely be the peak birth year if this had not occurred.

350 300 250 200 150 100 50 0

289

117 8

1

3

4

7

26 26 3 18

49 53

146

175 158

205 213

268 284

290

267

226

218

184 167 127 90

76

4

19 68 19 69 19 70 19 71 19 72 19 73 19 74 19 75 19 76 19 77 19 78 19 79 19 80 19 81 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97

Number of respondents

Figure 1: Respondent birth years

Year

Figure 1 shows a small and slow increase in the number of respondents born from 1968 to 1979. The vast majority of respondents were born in 1980 or later. Gender Figure 2 shows that approximately three times as many women took the survey as men.

6 Figure 2: Respondent gender Other gender 2% Male (cis) 26%

Female (cis) 72%

72% of respondents were cisgender females, while only 26% were cisgender males. 2% of those surveyed were intersex, transgender, genderqueer, or identified as another gender. There are several plausible explanations for why more women than men responded to the survey. Girls tend to have more positive attitudes towards education than boys, so they would be more likely to identify strongly with their educational background. Christian fundamentalism (see page 21) is particularly concerned with the subordinate status of girls, and the majority of the homeschool abuse survivor community (where the survey originated) is female. It is possible that homeschooling is gendered feminine; however, it is unclear which, if any, of these reasons influenced the gender disparity in respondents. Race As shown in Figure 3, the survey sample was around 85% ‘White’. The other biggest racial or ethnic categories were ‘White/Native American’ (3% of the sample), multiple races (2%), ‘White/Latino’ (2%), ‘Other’ (4%), and ‘Latino’ (2%).

7 Figure 3: Race of respondents

White White/Native American

112 73 72 144

Mixed race White/Latino Latino

61

154

30

White/East Asian East Asian Native American

27 3148

17

African American Other Races Other

13 6

NCES data shows that the survey overrepresents white alumni and underrepresents of black and Latino alumni; homeschoolers in 2011 were about 68% ‘White’, 8% ‘Black’, and 15% ‘Hispanic’.

Class Approximately one fifth of respondents (19.2%) reported that they did not know their family’s income bracket during the time they were homeschooled. The rest of the respondents are represented in Figure 4, which shows a disproportionate number of respondents (10.2% of the sample) reporting that their families made more than $100,000 per year.

500 450 400 350 300 250 200 150 100 50 0

430

425 376

377 302

289

266 188

194

115

$9 0,

90 ,0

00 00 010 0, 00 0 $1 00 ,0 00 +

00 00 0$8 0,

00 0$7 0,

$6 0,

00 0-

70 ,0

80 ,0

00

00 60 ,0

00 $5 0,

00 0-

50 ,0

00 00 0$4 0,

$3 0,

00 0-

40 ,0

30 ,0

00 00 0-

20 ,0 $2 0,

00 0-

$1 0,

$0 -1 0

00

29

,0 00


Figure 4: What bracket best represents your family's average annual household income during the time you were homeschooled?

Note, however, that these income brackets do not take into account the number of children each household had (and thus the real value of the income), the year the respondent was born (and therefore the comparison with today’s dollars), or whether the family underwent a significant change in income during the respondent’s homeschooling—all factors which could result in erroneous comparisons between responses. Further, this data also relies solely on self‐reports, which may be inaccurate due to lack of knowledge of one’s parents’ finances.

8 Location The survey had respondents from all 50 US states, plus DC and Puerto Rico. Figure 5 shows that in most states, the number of respondents was under 5% of the total (that is, less than 166 respondents). However, California and Texas each accounted for between 9% and 10% of respondents. The main concentrations of respondents were in the Midwest and on the Pacific coast. As we do not know how many homeschoolers there are in each state, it is impossible to tell how representative the sample was in terms of location; however, it seems possible that California and Texas are overrepresented. Figure 5: Respondent locations

A further 10% of the sample was homeschooled overseas or in multiple locations.

Siblings The respondents’ mean number of siblings was 4 and the median was 3. Figure 6 shows a peak in the number of siblings at 2 and 3, followed by a slow decline. The vast majority of respondents had between 1 and 6 siblings; homeschool alumni in our sample were unlikely to be only children.


Figure 6: Respondents' number of siblings 800 700 600 500 400 300 200 100 0

726

720

544 457 350 248 179 93

0

1

2

3

4

5

6

7

153

8

92

9

52

35

14

10

7

4

3

2

0

1

10

11

12

13

14

15

16

17

18

19

Number of siblings

9 Note that this figure does not include the one respondent who had more than 19 siblings. 2. Sexuality This section includes a description of respondents’ sexual orientations, change in personal attitude toward the GLBTQ* community, and personal attitude toward kink/BDSM. Orientation As shown in Figure 7, 83% of respondents identified as ‘Straight’. This means that around one in five respondents had a sexual orientation other than ‘Straight’. The most common of these was ‘Bisexual’ at 7% of the total sample. Each group of ‘Gay’, ‘Asexual’, ‘Pansexual’, and ‘Lesbian’ respondents composed 1‐2% of the total sample. Around 1‐2% of respondents also identified as both ‘Bisexual’ and ‘Pansexual’; 1‐2% identified as both ‘Straight’ and ‘Asexual; and 1‐2% identified as both ‘Straight’ and either ‘Bisexual’ or ‘Pansexual’ or both. The rest of the respondents either identified as multiple orientations or as ‘Other’. Figure 7: Respondents' sexual orientation

Straight/Bi/Pan, 42 Pansexual, 43 Asexual, 47 Straight, 3058

Gay, 56

Lesbian, 39 Straight/Asexual, 31 Bi/Pan, 26 Multiple orientations, 28 Other, 58 Bisexual, 274

Personal attitude towards GLBTQ* Overall, respondents’ opinions of the GLBTQ* community changed dramatically from childhood to the present, from moderately negative to moderately positive. As shown in Figure 8, respondents’ attitude towards this group in childhood was, on average, 2.25—closer to ‘Negative’ than ‘Neutral’. At present, however, respondents’ attitudes average 3.42, halfway between ‘Neutral’ and ‘Positive’—a more than 150% increase.

10 Figure 8: Change in respondents' attitude towards GLBTQ* persons from childhood to the present 1705

1800


1600 1400

1306 1130

1200 957

1000

Then

800

673

617

Now

663

600 400 200

62

150

143

0 Very positive

Positive

Neutral

Negative

Very negative

Personal attitude towards kink/BDSM In rating how strongly they agreed with the statement ‘I consider BDSM/kink to be an important part of my identity’, responses averaged to 1.49, halfway between ‘Strongly disagree’ and ‘Disagree’. Figure 9: I consider BDSM/kink to be an important part of my identity.


3000

2793

2500 2000 1500 1000

715 315

500

80

49

Agree

Strongly agree

0 Strongly disagree

Disagree

Neutral

Figure 9 shows that 95% of respondents disagreed with the statement while 3% agreed.

3. Family & work This section includes a description of respondents’ relationship status, number of children and intent to have children, preferred method for educating children, employment status, and personal attitude toward homeschooling. Relationship status Almost half of the respondents were married and one third were not in a relationship, as shown in Figure 10. The rest were in a relationship or ‘Other’, which includes widows/widowers, divorced, separated, and remarried respondents, as well as those who belong to more than one category.

11 Figure 10: Respondents' marital status

In casual relationship(s) 3%

Other 6%

Engaged 3%

In serious relationship(s) 12%

Not in a relationship 30%

Married 46%

Children About one‐third of respondents have children, as shown in Figure 11. Note that half of respondents are under age 26, and many millennials are pushing back having children until their late 20s or early 30s; this number may not be representative of how many respondents will eventually have children.

12 Figure 11: Do you have children? Do you have children?

31.1%

No Yes

68.9%

The mean number of children respondents had was 3, as shown in Figure 12; the median was 2. These figures indicate that respondents have, on average, fewer children than they have siblings, although this may change as they grow older.


Figure 12: Number of respondents' children 500 400

386

356

300

216

200 92

100

36

17

7

2

2

1

Five

Six

Seven

Eight

Nine

Ten

37

0 One

Two

Three

Four

Other

Number of children

These numbers include stepchildren, foster children, and deceased children. The ‘Other’ category includes those expecting their first child and those who did not identify the number of children they have. Plans to have children Figure 13 shows that 42% of respondents reported that they were planning to have children, while around 31% said that they already had children. The remaining 27% of respondents (around one in four) were uncertain about their future choices or did not plan to have children. The most common reason respondents chose for not planning to have children was that they did not want them (10%).

13 Figure 13: If you do not have children, do you plan to? I do not plan to because I have never been in a relationship. 4% I have never been in a relationship + I do not want children. 1%

I do not plan to because I do not want children. 10% I have children. 31%

I do not plan to because I am unable to have children. 1% Multiple options chosen. 2%

Other 9%

I do not have children, but am planning to. 42%

Interestingly, when asked about educating their children, the survey participants responded differently; here only around 8% of the 3,702 respondents reported that they did not intend to have children, as shown in Figure 14. Figure 14: What is or would be your preferred method(s) of educating your children? I do not intend to have children 7.64%

92.36%

14 Preferred method for educating children At this point we consider the set of 1,262 respondents to this same question who a) have or intend to have children and b) did not choose ‘it depends’ or ‘more than one’ preferred method. This group is represented in Figure 15. The plurality of this group (47% of them) chose only homeschooling as their preferred method, with ‘only public school’ coming in second (19%) and ‘only private school’ being the least preferred method (10%). Figure 15: What is or would be your preferred method(s) of educating your children?

Of all 3,702 respondents, 35% (around one third) believed that their educational decisions should depend on each child, as shown in Figure 16. The other 65% stated that they would educate their children using one or more specific methods. Figure 16: What is or would be your preferred method(s) of educating your children?

It depends on each child. 35%

65%

15 Employment As shown in Figure 17 below, nearly three quarters of the 3,702 respondents (72%) are currently employed for pay, with two thirds of these (46% of the entire sample) employed full time. The unemployment rate for respondents in the sample was 6%, nearly identical to the national unemployment rate. Figure 17: Are you currently employed for pay? Not seeking paid work 9.2%

12.7%

Seeking paid work but not currently employed Employed part-time for pay

6.0%

Employed full-time for pay Other (please specify)

25.7%

46.3%

While 15% of respondents felt that they were not academically prepared for their current job (Figure 18), 60% did feel prepared. Figure 18: Do you feel that your homeschooling education gave you the academic skills you needed for your job? Yes

11.6%

No Not employed Not relevant to current job

13.1%

14.6%

60.7%

Personal attitude toward homeschooling When asked about their attitude toward homeschooling (Figure 19), respondents rated their experience on average at 2.38, closer to ‘Agree’ than to ‘Neutral’. About two‐thirds of respondents (62%) agreed with the statement, while about one‐quarter (25%) disagreed with the statement. The remaining respondents were neutral or ‘Other’.

16


Figure 19: My homeschool experience prepared me for the future. 1311

1400 1200

998

1000 800 600 400

498

414

364

117

200 0 Strongly agree

Agree

Neutral

Disagree

Strongly disagree

Other

4. Academics This section includes respondents’ ratings of their instruction in various subjects and their view of unschooling. K‐5 Academics In elementary education, participants rated all the subjects surveyed as better than neutral. The highest‐ ranked subject was reading, followed by other language arts topics. The lowest‐ranked subject was science, which still ranked better than neutral at 3.68 (see Figure 20; note that here 5 is the best rating and 1 is the worst). Figure 20: K‐5 Academics 5

4.76

Average score

4.39

4.38

4.26

4.16

4.10

4.09

4.05

4

3.68

3 2 1 Reading

Spelling

Grammar

Handwriting Technology use

Sex education

History

Arithmetic

Science

Subject

It is to be noted that, for all the subjects except for sex education and technology use, less than 3% of respondents were not offered the course. A full 40% of students were not offered sex education, and 19% were not offered technology use.

Secondary school academics In secondary education, participants rated all the subjects as better than neutral except for sex education. The only subjects that were rated over 4 (5 being the best) were English literature, writing/composition, and US history. In general, respondents tended to rate their experiences in humanities courses as better than those in the sciences, foreign language, and upper‐level math.

17 Figure 21: 6‐12 Academics English literature

4.23

Writing/composition

4.20

U.S. history

4.06

The arts

3.98

U.S. government/civics

3.97

World history

3.94

Communications (e.g., speech, debate)

3.87

Geography

3.86

World literature

3.85

Algebra

3.75

Geometry

3.69

Economics

3.69

Social science

3.67

Biology

3.65

Computer science

3.60

Trigonometry

3.58

Earth/space sciences

3.57

Chemistry

3.57

Foreign language(s)

3.51

Physics

3.48

Calculus

3.32

Statistics

3.25

Sex education

2.87 1

2

3

4

5

The sciences and higher‐level math were also the courses least likely to be offered by parents and the courses most likely to be skipped by students. These two variables also appear to be correlated, as shown in Figure 22 below.

18

2000 1800 1600 1400 1200 1000 800 600 400 200 0

Course not offered

Statistics

Calculus

Computer science

Sex education

Trigonometry

Physics

Economics

Social science

Communications

Chemistry

Earth sciences

The arts

Foreign language

World literature

Geometry

Geography

Biology

U.S. government

World history

English literature

U.S. history

Algebra

Course available but I chose to skip

Writing/composition


Figure 22: Secondary school 'Course not offered' vs. 'Course available but I chose to skip'

Unschooling A total of 547 respondents (15% of the sample) reported that they identified with the unschooling movement. The average rating of the question was 2.34, closer to ‘Disagree’ than to ‘Neutral’. Figure 23: I identify with the unschooling movement.


1200 1000 800 600 400 200 0 Strongly disagree

Disagree

Neutral

Agree

Strongly agree

5. Non‐academics This section includes respondents’ ratings of their instruction in various non‐academic subjects and information about their childcare duties, hobbies, sports, music, and view of medicine. Non‐academic subjects Respondents scored all non‐academic subjects at or above average (see Figure 24). The most highly ranked subjects were driver’s education and home economics; the lowest ranked subjects were diversity in religious and political studies.

19 Figure 24: Non‐academic subjects

Non-academic subjects 5

Average rating

4.14

4.14

4

3.73

3.66

3.61

3.61 3.03

3.00

Diversity in religious studies

Diversity in political studies

3

2

1 Driver's education

Home economics

College prep Socialization Job training, preparation for workplace

Physical education

A total of 75 respondents (2% of the sample) reported that socialization was not offered during their educations. Further, 261 respondents (7%) stated that college prep was not offered. Childcare duties One fifth of respondents had primary childcare or teaching duties for a sibling, as shown in Figure 25. Figure 25: Did you have primary childcare or teaching duties for a sibling? 20.0%

Yes No

80.0%

Hobbies Approximately one in eight respondents (12%) reported they were not allowed to have hobbies independent of their parents’ (see Figure 26).

20 Figure 26: Were you allowed to initiate hobbies based on your personal interests (independent from your parent(s)'s interests? 11.9%

Yes No

88.1%

Sports Around two thirds of respondents (62%) reported that they played sports, as shown in Figure 27. Figure 27: Did you do sports?

37.98% No Yes 62.02%

Music Nearly half of respondents reported they were not allowed to listen to contemporary music (Figure 28). Figure 28: Were you allowed to listen to contemporary music that was not explicitly 'Christian'?

46.2%

Yes 53.8%

No

Medicine Respondents’ attitudes towards doctors and medicine were positive in general. The mean score was 3.67, closer to “Positive” than to “Neutral”, and the median score was 4,response was “Positive”. (Figure 29)

21 Figure 29: What was your family’s general attitude towards mainstream doctors and hospitals? 1600

1422

Number of res pondent s

1400 1200 1000

868

840

800 600

450

400 122

200 0 Very positive

Positive

Neutral

Negative

Very negative

6. Religion This section includes respondents’ reports of the denomination they were raised in and whether it was fundamentalist, their attitude toward the religious beliefs of their upbringing, and their current religious beliefs. Denomination The plurality of respondents (28%) reported that the religious environment of their upbringing was not available in our response options (see Figure 30). The largest identifiable groups were Non‐Denominational (26%), followed by Baptist (19%), Presbyterian (6%), Reformed (5%), Independent Fundamental Baptist (4%), Assembly of God (3%), Catholic (3%), Calvinist (3%), and Charismatic (3%). Figure 30: What type of Christianity did your family practice for the majority of your homeschool experience?

Other/Unknown (please specify), 27.9%

Non-Denominational, 26.3%

Charismatic, 2.9% Calvinist, 3.0% Catholic, 3.2%

Baptist, 18.6%

Assembly of God, 3.2% Independent Fundamental Baptist, 4.0%

Presbyterian, 5.6%

Reformed, 5.3%

22 Fundamentalism “Christian Fundamentalism” was defined in the survey according to Marsden (1991) as a “militantly anti‐ modernist” form of “evangelicalism.” Christian Fundamentalism includes, but is not limited to, the following ideologies: Christian legalism, Quiverfull, young earth creationism, anti‐LGBT rights, Christian Patriarchy, modesty and purity culture, betrothal and/or courtship, stay‐at‐home daughter movement, Dominionism, and Christian Reconstructionism. It is not limited to Protestantism and can also be seen in Catholic, Mormon, and other subcultures. More than half of the survey respondents (58%) considered their religious upbringing to be fundamentalist (Figure 31). Figure 31: Would you consider the type of Christianity you were primarily raised with to be (more often than not) a form of Christian fundamentalism?

Yes

41.8%

No

58.2%

Attitude In general, homeschoolers had a positive attitude towards their religious upbringing, although some also had negative attitudes; few had neutral attitudes (see Figure 32).


Figure 32: The religious environment of my upbringing had a positive influence on my life. 1600 1400 1200 1000 800 600 400 200 0

1348 879

Strongly Agree

Agree

574

469

420

Neutral

Disagree

Strongly disagree

Figure 32 shows that respondents’ mean rating of their upbringing was 3.54, halfway between ‘Neutral’ and ‘Agree’. The median score was 4, ‘Agree’

23 Current religious beliefs Four out of five survey respondents still consider themselves Christian (Figure 33). Figure 33: Do you consider yourself a Christian today?

20.8%

Yes No

79.2%

Of those who do not identify as Christian, about one third (30%) identified as atheists, one third (31%) as agnostics, and the rest as having other beliefs. Figure 34: If you do not consider yourself a Christian today, what do you consider yourself? Other/Unknown, 17.72% Agnostic, 30.82% Pagan, 4.23% None, 4.89%

Spiritual but not religious, 11.90%

Atheist, 30.42%

7. Mental health This section includes respondents’ reports about visits to a therapist, diagnosed and undiagnosed mental illnesses, beliefs about parental mental illness, and self‐harm. Therapy Nearly half of all respondents (45%) have had an appointment with a therapist (Figure 35).

24 Figure 35: Have you ever had an appointment with a mental health counselor/therapist?

Yes No

44.5%

55.5%

Mental illness Figure 36 shows that 25% of respondents had been diagnosed with a mental illness by a mental health professional. Figure 36: Have you ever been diagnosed with a mental illness by a mental health professional?

Other 3% Diagnosed with one or more mental illnesses 25%

Never diagnosed with a mental illness 72%

If we also consider respondents who believe they have a mental illness but have never been diagnosed, the number of respondents with mental illnesses rises to 48% (Figure 37).

25 Figure 37: Do you have other mental health conditions which have never been diagnosed by a mental health professional?

Other 6%

Undiagnosed mental illness(es) 23% Never suffered from mental illness 46%

Diagnosed with one or more mental illnesses 25%

Figure 37 shows that 48% of respondents reported having a mental health condition.

Figure 38: Did your parent(s) suffer from any mental illness(es)? I don't know., 8%

No, 43%

I suspect so, but not diagnosed by a mental health professional, 35%

Yes, diagnosed by a mental health professional, 14%

Figure 38 shows that, while only 14% of respondents reported that their parent or caregiver had been diagnosed with a mental illness by a mental health professional, a further 35% believed their parent suffered from an undiagnosed mental illness.

26 Self‐harm More than a quarter of respondents (28%) report struggling with self‐injury (Figure 39). Figure 39: Have you ever struggled with self‐injury?

28%

Yes No

72%

Figure 40 shows that about 8% of respondents, or nearly 1 in 10, have attempted suicide. Figure 40: Have you ever attempted suicide? 8.2%

Yes No

91.8%

8. Abuse This section includes respondents’ reports of abuse, their attitude toward spanking, and changes in their attitude toward social services. Abuse survivors The majority of the respondents to this survey did not experience abuse—58% of respondents reported that they had not experienced abuse. The most frequently reported type of abuse in this survey was emotional abuse—almost one in three respondents reported experiencing emotional abuse. Figure 41 shows the

27 percentages for the types of abuse most frequently defined in state law (where ‘educational neglect’ refers to violations of the compulsory education statute and ‘economic abuse’ refers to violations of child labor laws). Figure 41: Frequency of types of abuse typically defined in state law Emotional abuse Educational neglect Physical abuse Economic abuse Sexual abuse Medical abuse

% of respondents who experienced this type of abuse 30.0% 17.0% 16.2% 8.9% 4.8% 4.3%

Figure 42 shows the comparative frequencies of respondents’ reports of abuse (note that some respondents reported more than one type of abuse, so the numbers do not add up to 3,702). The types of abuse are defined here. Figure 42: Respondents' experiences of abuse

Total number of respondents

3702

Did not experience abuse

2140

Emotional abuse

1111

Verbal abuse

1001

Religious abuse

856

Educational neglect

628

Physical abuse

598

Economic abuse

330

Educational abuse

226

Sexual abuse

178

Medical abuse

161

Sexual harassment

154

Identification abuse

135

Other (please specify)

130 0

500

1000

1500

2000

2500

3000

3500

4000

28 Spanking About a quarter of respondents (23%) consider spanking a child to be inherently abusive. (Figure 43) Figure 43: Do you consider spanking a child to be inherently abusive?

22.6%

Yes No

77.4%

Figure 44 shows that about one third of respondents (32%) stated that they would ‘Never’ spank their children. The mean value was 2.11, which falls closer to ‘Occasionally’ than to ‘Sometimes’. The median value was 2 ‘Occasionally’. Figure 44: I do/would spank my children.


1400 1200

1295 1181

1000

877

800 600 400

255

200

72

0 Never

Occasionally

Sometimes

Often

Always

Social services Figure 45 shows that overall, respondents’ opinions of various social services became much more positive in adulthood than they had been in childhood. On average, there is a 133% increase in positive attitudes toward social services.

29 Figure 45: Changes in respondents' impressions of social services, from childhood to the present

Impressions of respondents, from 'Strongly negative' (1) to 'Strongly positive' (5)

p

5

4

3.81

3.74

3.61

3.48 3.25

3

Then 2.77

2.62

2.6

2.7

2.77

Now

2

1 Mandatory reporters

Anonymous tips

Social workers

Foster care system

People working in the public schools

Conclusion In conclusion, this survey of 3,702 homeschooled adults reached mostly white women born in the late 80s and early 90s. Many respondents were raised in fundamentalist homes; most respondents are still Christian, married and have or intend to have children. Most are positive towards homeschooling and would homeschool their kids. The results of the survey do indicate that there are suffering children in the homeschool community whose needs need to be addressed. There are unusually high rates of GLBTQ* children and mental health issues, and emotional abuse, physical abuse, and educational neglect are much bigger problems than sexual abuse. In particular, science and math education and sexual education seem to be at a lower level than desirable. It is HARO’s mission to address these issues and help these children; we believe the homeschool movement can be revitalized from within.

30 Appendix A: Survey’s Introductory Page as seen by Respondents You are invited to participate in a survey conducted by Homeschool Alumni Reaching Out (HARO), an organization whose mission is to make homeschooling better for future generations through awareness and education, peer‐support networks, and resource development. Survey consultation was provided by the Coalition for Responsible Home Education (CRHE). The purpose of this survey is to investigate the life experiences of Christian homeschool alumni by collecting information that past surveys of homeschool alumni have not. This survey is open to anyone who was raised in a Christian homeschool environment, regardless of whether their experiences were positive or negative and regardless of their current religious beliefs. As we are not randomly sampling the population, our results will be descriptive rather than representative. We will use the information collected to advocate for the interests of current and future homeschooled children. The survey should take approximately 30 minutes to complete. Your participation in this survey is completely voluntary. There are no foreseeable risks associated with this project. However, this survey includes questions about sensitive personal experiences — including abuse and mental illness — that could be triggering to some people. If you feel uncomfortable answering any questions, you can withdraw from the survey at any point. You are welcome to take the survey at a later date. Your answers are not recorded until you click “Send Results” on the survey’s final page. Your survey responses will be strictly confidential and data from this research will be reported only in the aggregate. Your information will be coded, SSL encrypted, and will remain anonymous. Your IP address is not recorded. The results of this survey will be posted at Homeschoolers Anonymous. If you would like to be informed when the results are available, there are instructions for how to do so at the end of the survey. This process is kept separate from the survey to protect your anonymity. You will have until Monday, September 15, 2014 at 11:59 pm Pacific time to respond to the survey. Should you have any technical problems, questions about the survey, or if you feel your rights have been violated, please contact survey author R.L. Stollar here. If you have any questions about HARO or the Homeschoolers Anonymous project, please contact us at [email protected]. Thank you very much for your time and support. Please confirm your eligibility to participate (you must be able to say "Yes" to all 4 statements below) and start the survey by clicking on the Next button below. >I am 18 or older. >I was homeschooled for at least 7 years. >My homeschooling environment was classifiable as Christian. >This is my first time taking the survey through completion.