Descriptives

• Monoracial Cisgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  animal_same_aglow	1	71	1.296	0.5951	1	1.175
  animal_different_aglow	2	71	3.775	0.5126	4	3.895

  	mad	min	max	range	skew	kurtosis	se
  animal_same_aglow	0	1	4	3	2.217	5.334	0.07062
  animal_different_aglow	0	2	4	2	-2.17	3.834	0.06084

• Monoracial Transgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  animal_same_aglow	1	63	1.365	0.7471	1	1.196
  animal_different_aglow	2	63	3.746	0.5671	4	3.863

  	mad	min	max	range	skew	kurtosis	se
  animal_same_aglow	0	1	4	3	2.286	4.875	0.09412
  animal_different_aglow	0	1	4	3	-2.592	7.711	0.07144

• Multiracial Cisgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  animal_same_aglow	1	67	1.448	0.7025	1	1.309
  animal_different_aglow	2	67	3.791	0.4096	4	3.855

  	mad	min	max	range	skew	kurtosis	se
  animal_same_aglow	0	1	4	3	1.474	1.586	0.08582
  animal_different_aglow	0	3	4	1	-1.4	-0.0405	0.05004

Regression

DV = Responses to “same animal” and “different animal” trials scored with agreement with Feppy as lower. Fixed effects = participant group, trial type (same vs. different animal), age (continuous), participant group x trial type, participant group x age (continuous), trial type x age (continuous), participant group x trial type x age (continuous) Random effects = participant

Findings: There is no main effect of participant group, or in other words cisgender vs transgender and monoracial vs multiracial children showed no differences in the way they responded on the animal trials. There is a main effect of trial type such that participants were more likely to disagree with Feppy on the animal different trial, as we would predict and as seen in past work. There is also a main effect of age such that older participants were more likely to disagree with Feppy. Both main effects were qualified by a significant interaction between trial type and age. Simple slopes analysis revealed that, on the animal same trial, older participants were more likely to disagree with Feppy, though this relation disappeared on the animal different trial.

Monoracial cisgender and animal different are the reference groups.

## SIMPLE SLOPES ANALYSIS 
## 
## Slope of age_yrs when trial_type = animal_same_aglow: 
## 
##   Est.   S.E.   t val.      p
## ------ ------ -------- ------
##   0.09   0.03     3.57   0.00
## 
## Slope of age_yrs when trial_type = animal_different_aglow: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.00   0.03    -0.00   1.00

Individual response patterns:

Descriptive—“same animal” trial and “different animal” trial scored as agree or disagree and classified into four distinct response patterns: agree with Feppy on both trials, disagree with Feppy on both trials, agree on same trial and disagree on different trial (predicted response), disagree on same trial and agree on different trial. Counts and corresponding percentages of participants falling into each of the 4 groups reported.

Findings:
Of the four possible response patterns, the majority of participants (91%) showed the expected response pattern, agreeing with Feppy that two of the same kind of animal were the same but diagreeing with Feppy that two different kinds of animals were the same. 6.5% of participants disagreed with Feppy on both trials and 2.5% of participants agreed with Feppy on both trials. No participants showed the reverse of the expected pattern (agreeing on the animal different trial and disagreeing on the animal same trial).

Additionally, a chi-square test for goodness of fit was statistically significant, indicating that our observed frequencies were significantly different than the expected frequencies, or that the response pattern that participants showed was significantly different from random responding.

Descriptives

• Monoracial Cisgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  cis_same_aglow	1	71	1.577	0.9049	1	1.386
  cis_different_aglow	2	71	3.176	0.9146	3.5	3.307

  	mad	min	max	range	skew	kurtosis	se
  cis_same_aglow	0	1	4	3	1.483	1.163	0.1074
  cis_different_aglow	0.7413	1	4	3	-0.8228	-0.4694	0.1085

• Monoracial Transgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  cis_same_aglow	1	63	1.46	0.8949	1	1.235
  cis_different_aglow	2	63	2.802	1.127	3	2.873

  	mad	min	max	range	skew	kurtosis	se
  cis_same_aglow	0	1	4	3	1.911	2.499	0.1127
  cis_different_aglow	1.483	1	4	3	-0.4073	-1.353	0.142

• Multiracial Cisgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  cis_same_aglow	1	68	1.647	0.8243	1	1.518
  cis_different_aglow	2	67	3.082	0.9716	3	3.209

  	mad	min	max	range	skew	kurtosis	se
  cis_same_aglow	0	1	4	3	1.187	0.7809	0.09997
  cis_different_aglow	1.483	1	4	3	-0.8504	-0.3436	0.1187

Regression

DV = Responses to “same assigned sex, same gender identity” and “different assigned sex, different gender identity” trials scored with agreement with Feppy as lower. Fixed effects = participant group, trial type, age (continuous), participant group x trial type, participant group x age (continuous), trial type x age (continuous), participant group x trial type x age (continuous). Random effects = participant

Findings:
There is no main effect of participant group, or in other words cisgender vs transgender and monoracial vs multiracial children showed no differences in the way they responded on the gender trials from past work. There is a main effect of trial type such that participants were more likely to disagree with Feppy on the “different assigned sex, different gender identity” trial (i.e., two cisgender children of different genders), as we would predict and as seen in past work. There was not a main effect of age. There was a significant two-way interaction between participant group and trial type qualified by a significant three-way interaction between participant group, trial type, and age such that there was a relation between age and trial type for monoracial transgender participants and multiracial cisgender participants, with older participants from these two groups being more likely to disagree with Feppy that two cisgender children of the same gender were the same compared to monoracial cisgender participants.

Monoracial cisgender and cisgender different are the reference groups.

Individual response patterns:

Descriptive—“ same assigned sex, same gender identity” trial and “different assigned sex, different gender identity” trial scored as agree or disagree and classified into four distinct response patterns: agree with Feppy on both trials, disagree with Feppy on both trials, agree on same trial and disagree on different trial (predicted response), disagree on same trial and agree on different trial. Counts and corresponding percentages of participants falling into each of the 4 groups reported.

Findings:

Of the six possible response patterns, the majority of participants (51%) showed the expected response pattern, agreeing with Feppy that two cisgender kids of the same gender were the same but diagreeing with Feppy that two cisgender kids of different genders were the same. 21.3% of participants agreed with Feppy on both trials and 9.4% of participants disagreed with Feppy on both trials. Only 0.5% of participants showed the reverse of the expected pattern (agreeing on the cisgender different genders trial and disagreeing on the cisgender same genders trial). 14.9% of participants agreed on the cisgender same gender trial but showed mixed responding on the cisgender different genders trials (agreed on one but disagreed on the other), while 3% of participants disagreed on the cisgender same gender trial but showed mixed responding on the cisgender different genders trials (agreed on one but disagreed on the other).

Additionally, a chi-square test for goodness of fit was statistically significant, indicating that our observed frequencies were significantly different than the expected frequencies, or that the response pattern that participants showed was significantly different from random responding.

Descriptives

Regression

DV = composite (calculated by averaging across all four trials) with “same assigned sex, different gender identity” scored with agreement with Feppy as lower and “different assigned sex, same gender identity” scored with agreement with Feppy scored as lower. Fixed effects = participant group, age (continuous), participant group x age (continuous) interaction. Random effects = participant

Findings:
The intercept is 3.15, which suggests that, at baseline, children do tend to view transgender and cisgender children as qualitatively distinct, as this value can range from 1 to 4, with a higher score indicating greater view that transgender and cisgender children are qualitatively distinct. There is no main effect of participant group, or in other words cisgender vs transgender and monoracial vs multiracial children showed no differences in their tendency to view transgender and cisgender children as qualitatively distinct. There is also no main effect of age.

Monoracial cisgender is the reference group.

Individual response patterns

Descriptive—each of 2 “same assigned sex, different gender identity” trials scored as agree (0) or disagree (1), and each of 2 “different assigned sex, same gender identity” trials scored as agree (0) or disagree (1), summed together for a score from 0 to 4. Counts and corresponding percentages of participants falling into each of the 5 groups reported.

Findings:
Of the five possible response patterns, the most common response was a score of 2 (i.e., at the midpoint), with 37.3% of participants responding consistent with the idea that transgender and cisgender people are qualitatively distinct on two trials and inconsistent with the idea that transgender and cisgender people are qualitatively distinct on two trials. The next most common reponse was a score of 4, with 23.9% of participants responding consistent with the idea that transgender and cisgender people are qualitatively distinct on all trials, then 18.9% of participants responded consistent with the idea that transgender and cisgender people are qualitatively distinct on three of four trials. 11% of participants responded consistent with the idea that transgender and cisgender people are qualitatively distinct on one trial, and only 9% of participants responsed consistent with the idea that transgender and cisgender people are qualitatively distinct on zero trials.

Descriptives

Regression

DV = composite (calculated by averaging across all four trials) with “same assigned sex, different gender identity” scored with agreement with Feppy as higher and “different assigned sex, same gender identity” scored with agreement with Feppy scored as lower. Fixed effects = participant group, age (continuous), participant group x age (continuous) interaction. Random effects = participant.

Findings:
The intercept is 2.46, which suggests that, at baseline, children do not tend to view sex as determinative or gender identity as determinative, as this value can range from 1 to 4 and the intercept is almost right at the midpoint (a higher score indicating greater view that sex is determinitive and a lower score indicating greater view that gender identity is determinitive). There is no main effect of participant group, or in other words cisgender vs transgender and monoracial vs multiracial children showed no differences in their tendency to view sex or gender identity as determinitive. There is also no main effect of age.

Monoracial cisgender is the reference group.

Individual response patterns

Descriptive—each of 2 “same assigned sex, different gender identity” trials scored as agree (1) or disagree (0) and each of 2 “different assigned sex, same gender identity” trials scored as agree (0) or disagree (1) summed together for a score from 0 to 4. Counts and corresponding percentages of participants falling into each of the 5 groups reported.

Findings:
Of the five possible response patterns, the most common response was a score of 2 (i.e., at the midpoint), with 42.29% of participants responding consistent with the idea that sex is determinitive on two trials and consistent with the idea that gender identity is determinitive on two trials. The next most common reponse was a score of 0, with 25.9% of participants responding consistent with the idea that gender identity is determinitive on all trials, then 20.9% of participants responded consistent with the idea that gender identity is determinitive on three of four trials. 9% of participants responded consistent with the idea that sex is determinitive on three of four trials, and only 2% of participants responsed consistent with the idea that sex is determinitive on all trials.

Additionally, a chi-square test for goodness of fit was statistically significant, indicating that our observed frequencies were significantly different than the expected frequencies, or that the response pattern that participants showed was significantly different from random responding.

Descriptives

• Monoracial Cisgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  mono_same_aglow	1	71	2.239	1.048	2	2.175
  mono_different_aglow	2	71	2.796	1.009	3	2.868

  	mad	min	max	range	skew	kurtosis	se
  mono_same_aglow	1.483	1	4	3	0.2539	-1.199	0.1244
  mono_different_aglow	1.483	1	4	3	-0.3473	-1.093	0.1198

• Monoracial Transgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  mono_same_aglow	1	63	2.349	1.152	2	2.314
  mono_different_aglow	2	63	2.73	1.117	3	2.784

  	mad	min	max	range	skew	kurtosis	se
  mono_same_aglow	1.483	1	4	3	0.3017	-1.377	0.1452
  mono_different_aglow	1.483	1	4	3	-0.2855	-1.376	0.1408

• Multiracial Cisgender:

  Table continues below

  	vars	n	mean	sd	median	trimmed
  mono_same_aglow	1	67	2.507	0.9272	2	2.509
  mono_different_aglow	2	67	3.037	0.9467	3	3.155

  	mad	min	max	range	skew	kurtosis	se
  mono_same_aglow	1.483	1	4	3	0.2032	-0.9034	0.1133
  mono_different_aglow	1.483	1	4	3	-0.7519	-0.5115	0.1157

Regression

DV = Responses to “same race” and “different race” trials scored with agreement with Feppy as lower. Fixed effects = participant group, trial type, age (continuous), participant group x trial type, participant group x age (continuous), trial type x age (continuous), participant group x trial type x age (continuous). Random effects = participant

Findings:
There is no main effect of participant group, or in other words cisgender vs transgender and monoracial vs multiracial children showed no differences in the way they responded on the race trials from past work. There is a main effect of trial type such that participants were more likely to disagree with Feppy on the “different race” trials (i.e., monoracial targets of difference races), as we would predict and as seen in past work. There was a main effect of age, such that older participants were less likely to disagree with Feppy.

Monoracial cisgender and monoracial different are the reference groups.

Individual response patterns:

Descriptive—“ same race” trial and “different race” trial scored as agree or disagree and classified into four distinct response patterns: agree with Feppy on both trials, disagree with Feppy on both trials, agree on same trial and disagree on different trial (predicted response), disagree on same trial and agree on different trial. Counts and corresponding percentages of participants falling into each of the 4 groups reported.

Findings:

Of the six possible response patterns, the largest percentage of participants (35.9%) disagreed with Feppy on both the monoracial same race trial and the monoracial different race trials that they were the same kind of people, which would not be the expected response based on prior research. 28.4% of participants agreed on both the monoracial same race trial and the monoracial different race trials that they were the same kind of people. Only 20.9% of participants showed the expected response pattern, agreeing with Feppy that two monoracial kids of the same race were the same but diagreeing with Feppy that two monoracial kids of different races were the same. Only 1% of participants showed the reverse of the expected pattern (agreeing on the monoracial different race trials and disagreeing on the monoracial same race trial). 9.5% of participants agreed on the monoracial same race trial but showed mixed responding on the monoracial different race trials (agreed on one but disagreed on the other), while 4.5% of participants disagreed on the monoracial same race trial but showed mixed responding on the monoracial different race trials (agreed on one but disagreed on the other).

Additionally, a chi-square test for goodness of fit was statistically significant, indicating that our observed frequencies were significantly different than the expected frequencies, or that the response pattern that participants showed was significantly different from random responding.

Descriptives

Regression

DV = composite (calculated by averaging across all four trials) with “Multi + Black” scored with agreement with Feppy as lower and “Multi + White” scored with agreement with Feppy scored as lower. Fixed effects = participant group, age (continuous), participant group x age (continuous) interaction. Random effects = participant

Findings:
The intercept is 4.08, which suggests that, at baseline, children do tend to view multiracial and monoracial children as qualitatively distinct, as this value can range from 1 to 4 (Of note, the intercept is slightly higher than 4 so we may be extrapolating beyond our data), with a higher score indicating greater view that multiracial and monoracial children are qualitatively distinct. There is no main effect of participant group, or in other words cisgender vs transgender and monoracial vs multiracial children showed no differences in their tendency to view multiracial and monoracial children as qualitatively distinct. There is a main effect of age such that older children are less likely to view multiracial and monoracial children as qualitatively distinct.

Monoracial cisgender is the reference group.

Individual response patterns

Descriptive—each of 2 “Multi + Black” trials scored as agree (0) or disagree (1) and each of 2 “Multi + White” trials scored as agree (0) or disagree (1) summed together for a score from 0 to 4. Counts and corresponding percentages of participants falling into each of the 5 groups reported.

Findings:
Of the five possible response patterns, the most common response was a score of 4 (i.e., highest score), with 38.5% of participants responding consistent with the idea that multiracial and monoracial children are qualitatively distinct on all four trials. The next most common reponse was a score of 0, with 25.5% of participants responding consistent with the idea that multiracial and monoracial children are qualitatively distinct on zero trials, then 14.5% of participants responded consistent with the idea that multiracial and monoracial children are qualitatively distinct on three of four trials. 11% of participants responded consistent with the idea that multiracial and monoracial children are qualitatively distinct on two trials, and only 10.5% of participants responsed consistent with the idea that multiracial and monoracial children are qualitatively distinct on one trial.

Additionally, a chi-square test for goodness of fit was statistically significant, indicating that our observed frequencies were significantly different than the expected frequencies, or that the response pattern that participants showed was significantly different from random responding.

Descriptives

Regression

DV = composite (calculated by averaging across all four trials) with “Multi + Black” scored with agreement with Feppy as higher and “Multi + White” scored with agreement with Feppy scored as lower. Fixed effects = participant group, age (continuous), participant group x age (continuous) interaction.

Findings:
The intercept is 2.51, which suggests that, at baseline, children do not tend to engage in hypodescent or hyperdescent, as this value can range from 1 to 4 and the intercept is almost right at the midpoint (a higher score indicating greater hypodescent and a lower score indicating greater hyperdescent). There is no main effect of participant group, or in other words cisgender vs transgender and monoracial vs multiracial children showed no differences in their tendency to engage in hypodescent or hyperdescent. There is also no main effect of age.

Monoracial cisgender is the reference group.

## SIMPLE SLOPES ANALYSIS 
## 
## Slope of age_yrs when group_race_gender = Multiracial Cisgender: 
## 
##   Est.   S.E.   t val.      p
## ------ ------ -------- ------
##   0.04   0.02     1.77   0.08
## 
## Slope of age_yrs when group_race_gender = Monoracial Transgender: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.03   0.02    -1.35   0.18
## 
## Slope of age_yrs when group_race_gender = Monoracial Cisgender: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.01   0.02    -0.43   0.67

Individual response patterns

Descriptive—each of 2 “Multi + Black” trials scored as agree (1) or disagree (0) and each of 2 “Multi + White” trials scored as agree (0) or disagree (1) summed together for a score from 0 to 4. Counts and corresponding percentages of participants falling into each of the 5 groups reported.

Findings: Of the five possible response patterns, the most common response was a score of 2 (i.e., at the midpoint), with 71% of participants responding consistent with hypodescent on two trials and consistent with hyperdescent on two trials. The next most common response was a score of 1, with 15% of participants responding consistent with hypodescent on one trial, then 10% of participants responding consistent with hypodescent on three of four trials. 3% of participants responded consistent with hypodescent on zero trials, and only 1% of participants responsed consistent with hypodescent on all trials.

Additionally, a chi-square test for goodness of fit was statistically significant, indicating that our observed frequencies were significantly different than the expected frequencies, or that the response pattern that participants showed was significantly different from random responding.