RQ_2

This is what we know:

Males are associated with challenge and relevance
Interest is less predictive but predicts positive affect
Females report higher challenge when STEM-SB is supported
Creating products is more relevant when students have higher pre-intereset
When agency is supported and students have higher pre-interest, they repory higher positive affect

This is what we still need to figure out:

How robust are interactions?
Is using p < .05 to determine what interactions to focus on the best idea?

How this document is organized

The outcomes are numbered 1 - 4 (challenge, relevance, learning, and positive affect)
Within the section for each outcome, three models are specified (just main effects [basic skills, creating products, support for agency, and support for STEM-SB], main effects X gender interactions, and main effects X pre-interest interactions)

library(tidyverse)
library(lme4)
library(corrr)
library(jmRtools)
library(sjPlot)

esm <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-esm.csv")
pre_survey_data_processed <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-pre-survey.csv")
post_survey_data_partially_processed <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-post-survey.csv")
video <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-video.csv")
pqa <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-pqa.csv")
attendance <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-attendance.csv")
class_data <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-class-video.csv")
demographics <- read_csv("/Volumes/SCHMIDTLAB/PSE/data/STEM-IE/STEM-IE-demographics.csv")
pm <- read_csv("/Volumes/SCHMIDTLAB/PSE/Data/STEM-IE/STEM-IE-program-match.csv")

load("~/Desktop/all-data.Rdata")

attendance <- rename(attendance, participant_ID = ParticipantID)
attendance <- mutate(attendance, prop_attend = DaysAttended / DaysScheduled, 
                     participant_ID = as.integer(participant_ID))
attendance <- select(attendance, participant_ID, prop_attend)

demographics <- filter(demographics, participant_ID!= 7187)
demographics <- left_join(demographics, attendance)

esm$overall_engagement <- jmRtools::composite_mean_maker(esm, hard_working, concentrating, enjoy, interest)

df <- left_join(esm, pre_survey_data_processed, by = "participant_ID") # df & post-survey
df <- left_join(df, video, by = c("program_ID", "response_date", "sociedad_class", "signal_number")) # df & video
df <- left_join(df, demographics, by = c("participant_ID", "program_ID")) # df and demographics

pqa <- mutate(pqa, 
              active = active_part_1 + active_part_2,
              ho_thinking = ho_thinking_1 + ho_thinking_2 + ho_thinking_3,
              belonging = belonging_1 + belonging_2,
              agency = agency_1 + agency_2 + agency_3 + agency_4,
              youth_development_overall = active_part_1 + active_part_2 + ho_thinking_1 + ho_thinking_2 + ho_thinking_3 + belonging_1 + belonging_2 + agency_1 + agency_2 + agency_3 + agency_4,
              making_observations = stem_sb_8,
              data_modeling = stem_sb_2 + stem_sb_3 + stem_sb_9,
              interpreting_communicating = stem_sb_6,
              generating_data = stem_sb_4,
              asking_questions = stem_sb_1)

# pqa <- rename(pqa, sixth_math_sociedad = sixth_math)
# pqa <- rename(pqa, seventh_math_sociedad = seventh_math)
# pqa <- rename(pqa, eighth_math_sociedad = eighth_math)
# pqa <- rename(pqa, dance_sociedad = dance)
# pqa <- rename(pqa, robotics_sociedad = robotics)

pqa$sociedad_class <- ifelse(pqa$eighth_math == 1, "8th Math",
                             ifelse(pqa$seventh_math == 1, "7th Math",
                                    ifelse(pqa$sixth_math == 1, "6th Math",
                                           ifelse(pqa$robotics == 1, "Robotics",
                                                  ifelse(pqa$dance == 1, "Dance", NA)))))

pqa <- rename(pqa, 
              program_ID = SiteIDNumeric,
              response_date = resp_date,
              signal_number = signal)

pqa$program_ID <- as.character(pqa$program_ID)

df <- left_join(df, pqa, by = c("response_date", "program_ID", "signal_number", "sociedad_class"))

df <- df %>% 
    mutate(youth_activity_three = case_when(
        youth_activity_rc == "Creating Product" ~ "Creating Product",
        youth_activity_rc == "Basic Skills Activity" ~ "Basic Skills Activity",
        TRUE ~ "Other"
    ))

df$youth_activity_three <- fct_relevel(df$youth_activity_three, 
                                       "Other")

df <- df %>% 
    mutate(ho_thinking_dummy = ifelse(sum_ho_thinking > 0, 1, 0),
           agency_dummy = ifelse(sum_agency > 0, 1, 0),
           active_dummy = ifelse(sum_ap > 0, 1, 0),
           belonging_dummy = ifelse(sum_belonging > 0, 1, 0),
           stem_sb_dummy = ifelse(sum_stem_sb > 0, 1, 0))

Correlations

df %>% 
    select(sum_agency, sum_stem_sb, overall_pre_interest) %>% 
    correlate() %>% 
    shave() %>% 
    fashion() %>% 
    knitr::kable()

rowname	sum_agency	sum_stem_sb
sum_agency
sum_stem_sb	.39
overall_pre_interest	.08	.02

df$gender <- as.factor(df$gender)

1. Challenge

1A: Just main effects of select activities / STEM-SB and pre interest and gender

m <- lmer(challenge ~ 
              youth_activity_three +
              agency_dummy +
              stem_sb_dummy + 
              
              overall_pre_interest + 
              gender + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	challenge
	B	std. Error	p
Fixed Parts
(Intercept)	1.98	0.24	<.001
youth_activity_three (Basic Skills Activity)	0.13	0.06	.034
youth_activity_three (Creating Product)	0.33	0.07	<.001
agency_dummy	0.08	0.06	.206
stem_sb_dummy	0.02	0.07	.795
overall_pre_interest	0.01	0.07	.936
gender (M)	0.25	0.11	.027
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.049
ICC_{participant_ID}	0.379
ICC_{program_ID}	0.042
Observations	2549
R² / Ω₀²	.530 / .523

1B: Adding interactions with gender

m <- lmer(challenge ~ 
              youth_activity_three*gender +
              agency_dummy*gender +
              stem_sb_dummy*gender + 
              overall_pre_interest + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	challenge
	B	std. Error	p
Fixed Parts
(Intercept)	1.86	0.25	<.001
youth_activity_threeBasic Skills Activity	0.15	0.07	.045
youth_activity_threeCreating Product	0.33	0.08	<.001
genderM	0.52	0.15	<.001
agency_dummy	0.11	0.08	.150
stem_sb_dummy	0.13	0.08	.107
overall_pre_interest	0.01	0.07	.927
youth_activity_threeBasic Skills Activity:genderM	-0.03	0.09	.707
youth_activity_threeCreating Product:genderM	0.01	0.10	.880
genderM:agency_dummy	-0.08	0.09	.408
genderM:stem_sb_dummy	-0.25	0.10	.012
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.048
ICC_{participant_ID}	0.380
ICC_{program_ID}	0.042
Observations	2549
R² / Ω₀²	.531 / .524

sjp.int(m)

1C: Adding interactions with pre-interest

m <- lmer(challenge ~ 
              youth_activity_three*overall_pre_interest + 
              agency_dummy*overall_pre_interest +
              stem_sb_dummy*overall_pre_interest + 
              gender +
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	challenge
	B	std. Error	p
Fixed Parts
(Intercept)	1.66	0.29	<.001
youth_activity_threeBasic Skills Activity	0.18	0.17	.278
youth_activity_threeCreating Product	0.50	0.20	.015
overall_pre_interest	0.12	0.09	.195
agency_dummy	0.21	0.18	.232
stem_sb_dummy	0.23	0.19	.214
genderM	0.25	0.11	.026
youth_activity_threeBasic Skills Activity:overall_pre_interest	-0.01	0.05	.782
youth_activity_threeCreating Product:overall_pre_interest	-0.05	0.06	.404
overall_pre_interest:agency_dummy	-0.05	0.05	.400
overall_pre_interest:stem_sb_dummy	-0.08	0.06	.201
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.051
ICC_{participant_ID}	0.379
ICC_{program_ID}	0.042
Observations	2549
R² / Ω₀²	.532 / .525

sjp.int(m)

2. Affect

2A: Just main effects of select activities / STEM-SB and pre interest and gender

m <- lmer(relevance ~ 
              youth_activity_three +
              agency_dummy +
              stem_sb_dummy + 
              
              overall_pre_interest + 
              gender + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	relevance
	B	std. Error	p
Fixed Parts
(Intercept)	2.10	0.21	<.001
youth_activity_three (Basic Skills Activity)	0.04	0.04	.269
youth_activity_three (Creating Product)	0.12	0.04	.004
agency_dummy	-0.05	0.04	.187
stem_sb_dummy	0.11	0.04	.011
overall_pre_interest	0.10	0.06	.125
gender (M)	0.24	0.11	.030
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.012
ICC_{participant_ID}	0.519
ICC_{program_ID}	0.011
Observations	2549
R² / Ω₀²	.594 / .592

2B: Adding interactions with gender

m <- lmer(relevance ~ 
              youth_activity_three*gender +
              agency_dummy*gender +
              stem_sb_dummy*gender + 
              overall_pre_interest + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	relevance
	B	std. Error	p
Fixed Parts
(Intercept)	2.03	0.21	<.001
youth_activity_threeBasic Skills Activity	0.08	0.05	.124
youth_activity_threeCreating Product	0.18	0.06	.002
genderM	0.37	0.13	.006
agency_dummy	-0.05	0.05	.323
stem_sb_dummy	0.16	0.05	.004
overall_pre_interest	0.10	0.06	.125
youth_activity_threeBasic Skills Activity:genderM	-0.07	0.07	.277
youth_activity_threeCreating Product:genderM	-0.11	0.08	.154
genderM:agency_dummy	-0.01	0.07	.891
genderM:stem_sb_dummy	-0.11	0.07	.144
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.012
ICC_{participant_ID}	0.520
ICC_{program_ID}	0.011
Observations	2549
R² / Ω₀²	.595 / .592

sjp.int(m)

2C: Adding interactions with pre-interest

m <- lmer(relevance ~ 
              youth_activity_three*overall_pre_interest + 
              agency_dummy*overall_pre_interest +
              stem_sb_dummy*overall_pre_interest + 
              gender +
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	relevance
	B	std. Error	p
Fixed Parts
(Intercept)	2.11	0.24	<.001
youth_activity_threeBasic Skills Activity	0.04	0.12	.727
youth_activity_threeCreating Product	-0.24	0.15	.122
overall_pre_interest	0.09	0.08	.217
agency_dummy	-0.03	0.13	.790
stem_sb_dummy	0.15	0.14	.287
genderM	0.24	0.11	.028
youth_activity_threeBasic Skills Activity:overall_pre_interest	0.00	0.04	.969
youth_activity_threeCreating Product:overall_pre_interest	0.11	0.05	.016
overall_pre_interest:agency_dummy	-0.01	0.04	.865
overall_pre_interest:stem_sb_dummy	-0.01	0.04	.779
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.013
ICC_{participant_ID}	0.518
ICC_{program_ID}	0.012
Observations	2549
R² / Ω₀²	.595 / .593

sjp.int(m)

3. Affect

3A: Just main effects of select activities / STEM-SB and pre interest and gender

m <- lmer(learning ~ 
              youth_activity_three +
              agency_dummy +
              stem_sb_dummy + 
              
              overall_pre_interest + 
              gender + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	learning
	B	std. Error	p
Fixed Parts
(Intercept)	2.35	0.19	<.001
youth_activity_three (Basic Skills Activity)	0.15	0.05	.003
youth_activity_three (Creating Product)	0.04	0.05	.435
agency_dummy	0.03	0.05	.578
stem_sb_dummy	0.10	0.05	.054
overall_pre_interest	0.08	0.06	.171
gender (M)	0.07	0.10	.522
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.008
ICC_{participant_ID}	0.354
ICC_{program_ID}	0.000
Observations	2548
R² / Ω₀²	.417 / .411

3B: Adding interactions with gender

m <- lmer(learning ~ 
              youth_activity_three*gender +
              agency_dummy*gender +
              stem_sb_dummy*gender + 
              overall_pre_interest + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	learning
	B	std. Error	p
Fixed Parts
(Intercept)	2.27	0.20	<.001
youth_activity_threeBasic Skills Activity	0.21	0.06	.001
youth_activity_threeCreating Product	0.06	0.07	.402
genderM	0.23	0.14	.118
agency_dummy	0.08	0.07	.254
stem_sb_dummy	0.12	0.07	.080
overall_pre_interest	0.08	0.06	.160
youth_activity_threeBasic Skills Activity:genderM	-0.14	0.09	.118
youth_activity_threeCreating Product:genderM	-0.04	0.10	.684
genderM:agency_dummy	-0.10	0.09	.268
genderM:stem_sb_dummy	-0.05	0.10	.618
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.008
ICC_{participant_ID}	0.355
ICC_{program_ID}	0.000
Observations	2548
R² / Ω₀²	.418 / .412

sjp.int(m)

3C: Adding interactions with pre-interest

m <- lmer(learning ~ 
              youth_activity_three*overall_pre_interest + 
              agency_dummy*overall_pre_interest +
              stem_sb_dummy*overall_pre_interest + 
              gender +
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	learning
	B	std. Error	p
Fixed Parts
(Intercept)	2.29	0.25	<.001
youth_activity_threeBasic Skills Activity	0.26	0.15	.100
youth_activity_threeCreating Product	-0.00	0.20	.991
overall_pre_interest	0.10	0.08	.207
agency_dummy	-0.08	0.16	.632
stem_sb_dummy	0.26	0.18	.146
genderM	0.07	0.10	.516
youth_activity_threeBasic Skills Activity:overall_pre_interest	-0.04	0.05	.475
youth_activity_threeCreating Product:overall_pre_interest	0.01	0.06	.839
overall_pre_interest:agency_dummy	0.04	0.05	.501
overall_pre_interest:stem_sb_dummy	-0.05	0.06	.375
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.009
ICC_{participant_ID}	0.354
ICC_{program_ID}	0.000
Observations	2548
R² / Ω₀²	.418 / .412

sjp.int(m)

4. Affect

4A: Just main effects of select activities / STEM-SB and pre interest and gender

m <- lmer(positive_affect ~ 
              youth_activity_three +
              agency_dummy +
              stem_sb_dummy + 
              
              overall_pre_interest + 
              gender + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	positive_affect
	B	std. Error	p
Fixed Parts
(Intercept)	1.94	0.25	<.001
youth_activity_three (Basic Skills Activity)	0.02	0.05	.748
youth_activity_three (Creating Product)	-0.05	0.05	.360
agency_dummy	0.17	0.05	.002
stem_sb_dummy	-0.04	0.05	.422
overall_pre_interest	0.20	0.07	.007
gender (M)	0.16	0.11	.160
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.027
ICC_{participant_ID}	0.432
ICC_{program_ID}	0.074
Observations	2549
R² / Ω₀²	.587 / .583

4B: Adding interactions with gender

m <- lmer(positive_affect ~ 
              youth_activity_three*gender +
              agency_dummy*gender +
              stem_sb_dummy*gender + 
              overall_pre_interest + 
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	positive_affect
	B	std. Error	p
Fixed Parts
(Intercept)	1.92	0.26	<.001
youth_activity_threeBasic Skills Activity	0.02	0.06	.771
youth_activity_threeCreating Product	-0.10	0.07	.133
genderM	0.20	0.14	.173
agency_dummy	0.20	0.06	.003
stem_sb_dummy	-0.04	0.07	.509
overall_pre_interest	0.20	0.07	.006
youth_activity_threeBasic Skills Activity:genderM	0.00	0.08	.991
youth_activity_threeCreating Product:genderM	0.11	0.09	.199
genderM:agency_dummy	-0.07	0.08	.391
genderM:stem_sb_dummy	-0.00	0.09	.980
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.027
ICC_{participant_ID}	0.431
ICC_{program_ID}	0.074
Observations	2549
R² / Ω₀²	.587 / .583

sjp.int(m)

4C: Adding interactions with pre-interest

m <- lmer(positive_affect ~ 
              youth_activity_three*overall_pre_interest + 
              agency_dummy*overall_pre_interest +
              stem_sb_dummy*overall_pre_interest + 
              gender +
              (1|program_ID) + (1|participant_ID) + (1|beep_ID_new),
          data = df)

sjPlot::sjt.lmer(m, p.kr = T, show.re.var = F, show.ci = F, show.se = T)

	positive_affect
	B	std. Error	p
Fixed Parts
(Intercept)	1.96	0.29	<.001
youth_activity_threeBasic Skills Activity	0.04	0.15	.802
youth_activity_threeCreating Product	-0.23	0.18	.207
overall_pre_interest	0.19	0.09	.032
agency_dummy	-0.12	0.15	.417
stem_sb_dummy	0.24	0.16	.149
genderM	0.16	0.11	.162
youth_activity_threeBasic Skills Activity:overall_pre_interest	-0.01	0.05	.887
youth_activity_threeCreating Product:overall_pre_interest	0.05	0.05	.323
overall_pre_interest:agency_dummy	0.10	0.05	.047
overall_pre_interest:stem_sb_dummy	-0.09	0.05	.086
Random Parts
N_{beep_ID_new}	236
N_{participant_ID}	180
N_{program_ID}	9
ICC_{beep_ID_new}	0.026
ICC_{participant_ID}	0.432
ICC_{program_ID}	0.074
Observations	2549
R² / Ω₀²	.587 / .583

sjp.int(m)

RQ_2

Joshua Rosenberg

12/19/2017

RQ #2: How is the relationship between STEM practices and perceptions of challenge, relevance, learning, and affect moderated by youth characteristics at program entry?

This is what we know:

This is what we still need to figure out:

How this document is organized

Correlations

1. Challenge

1A: Just main effects of select activities / STEM-SB and pre interest and gender

1B: Adding interactions with gender

1C: Adding interactions with pre-interest

2. Affect

2A: Just main effects of select activities / STEM-SB and pre interest and gender

2B: Adding interactions with gender

2C: Adding interactions with pre-interest

3. Affect

3A: Just main effects of select activities / STEM-SB and pre interest and gender

3B: Adding interactions with gender

3C: Adding interactions with pre-interest

4. Affect

4A: Just main effects of select activities / STEM-SB and pre interest and gender

4B: Adding interactions with gender

4C: Adding interactions with pre-interest