library(afex) # to run the ANOVA and plot results
library(psych) # for the describe() command
library(ggplot2) # to visualize our results
library(expss) # for the cross_cases() command
library(car) # for the leveneTest() command
library(emmeans) # for posthoc tests
library(effsize) # for the cohen.d() command
library(apaTables) # to create our correlation table
library(kableExtra) # to create our correlation table
library(sjPlot) # to visualize our resultsAI Experiment Analysis
Loading Libraries
Importing Data
# # import your AI results dataset
d <- read.csv(file="Data/myfinalresults.csv", header=T)State Your Hypotheses & Chosen Tests
H1: Participants in the social media use condition will have higher post-stress scores than those in the control condition. My chosen test for this hypothesis is an independent samples t-test. The dependent variable will be the post-stress score, and the independent variable will be the condition (social media use vs. control).
H2: Pre-survey stress scores will be positively correlated with post-survey stress scores across all participants. I will be using a Pearson’s correlation test for this hypothesis, using pre-survey and post-survey stress scores as continuous variables.
Check Your Variables
This is just basic variable checking that is used across all HW assignments.
# # to view stats for all variables
describe(d) vars n mean sd median trimmed mad min max range skew
id 1 100 50.50 29.01 50.50 50.50 37.06 1.00 100.00 99.0 0.00
identity* 2 100 50.50 29.01 50.50 50.50 37.06 1.00 100.00 99.0 0.00
consent* 3 100 1.93 0.26 2.00 2.00 0.00 1.00 2.00 1.0 -3.32
age 4 100 29.08 7.01 28.00 27.85 2.97 18.00 65.00 47.0 2.50
race 5 100 4.57 1.59 6.00 4.68 1.48 1.00 7.00 6.0 -0.28
gender 6 100 2.17 0.84 2.00 2.00 0.00 1.00 7.00 6.0 3.70
manip_out* 7 100 50.50 29.01 50.50 50.50 37.06 1.00 100.00 99.0 0.00
survey1 8 100 3.09 0.43 3.18 3.12 0.47 2.23 3.63 1.4 -0.45
survey2 9 100 3.38 0.20 3.50 3.42 0.00 2.50 3.60 1.1 -1.82
ai_manip* 10 100 50.50 29.01 50.50 50.50 37.06 1.00 100.00 99.0 0.00
condition 11 100 1.50 0.50 1.50 1.50 0.74 1.00 2.00 1.0 0.00
kurtosis se
id -1.24 2.90
identity* -1.24 2.90
consent* 9.11 0.03
age 8.27 0.70
race -1.54 0.16
gender 14.36 0.08
manip_out* -1.24 2.90
survey1 -1.37 0.04
survey2 3.50 0.02
ai_manip* -1.24 2.90
condition -2.02 0.05#
# # we'll use the describeBy() command to view skew and kurtosis across our IVs
describeBy(d, group = "survey1")
Descriptive statistics by group
survey1: 2.23
vars n mean sd median trimmed mad min max range skew kurtosis se
id 1 1 53.00 NA 53.00 53.00 0 53.00 53.00 0 NA NA NA
identity 2 1 83.00 NA 83.00 83.00 0 83.00 83.00 0 NA NA NA
consent 3 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA
age 4 1 42.00 NA 42.00 42.00 0 42.00 42.00 0 NA NA NA
race 5 1 6.00 NA 6.00 6.00 0 6.00 6.00 0 NA NA NA
gender 6 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA
manip_out 7 1 53.00 NA 53.00 53.00 0 53.00 53.00 0 NA NA NA
survey1 8 1 2.23 NA 2.23 2.23 0 2.23 2.23 0 NA NA NA
survey2 9 1 3.50 NA 3.50 3.50 0 3.50 3.50 0 NA NA NA
ai_manip 10 1 59.00 NA 59.00 59.00 0 59.00 59.00 0 NA NA NA
condition 11 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA
------------------------------------------------------------
survey1: 2.27
vars n mean sd median trimmed mad min max range skew kurtosis se
id 1 1 1.00 NA 1.00 1.00 0 1.00 1.00 0 NA NA NA
identity 2 1 96.00 NA 96.00 96.00 0 96.00 96.00 0 NA NA NA
consent 3 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA
age 4 1 29.00 NA 29.00 29.00 0 29.00 29.00 0 NA NA NA
race 5 1 6.00 NA 6.00 6.00 0 6.00 6.00 0 NA NA NA
gender 6 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA
manip_out 7 1 22.00 NA 22.00 22.00 0 22.00 22.00 0 NA NA NA
survey1 8 1 2.27 NA 2.27 2.27 0 2.27 2.27 0 NA NA NA
survey2 9 1 3.50 NA 3.50 3.50 0 3.50 3.50 0 NA NA NA
ai_manip 10 1 56.00 NA 56.00 56.00 0 56.00 56.00 0 NA NA NA
condition 11 1 1.00 NA 1.00 1.00 0 1.00 1.00 0 NA NA NA
------------------------------------------------------------
survey1: 2.5
vars n mean sd median trimmed mad min max range skew
id 1 25 50.48 30.42 59.0 50.48 38.55 2.0 99.0 97.0 -0.14
identity 2 25 50.12 32.35 54.0 50.14 38.55 1.0 99.0 98.0 -0.04
consent 3 25 1.92 0.28 2.0 2.00 0.00 1.0 2.0 1.0 -2.91
age 4 25 28.88 5.26 28.0 28.81 5.93 18.0 39.0 21.0 0.17
race 5 25 5.64 1.04 6.0 5.90 0.00 2.0 6.0 4.0 -2.52
gender 6 25 2.04 0.68 2.0 2.00 0.00 1.0 5.0 4.0 3.07
manip_out 7 25 49.00 29.10 50.0 49.43 43.00 1.0 93.0 92.0 -0.15
survey1 8 25 2.50 0.00 2.5 2.50 0.00 2.5 2.5 0.0 NaN
survey2 9 25 3.40 0.20 3.5 3.43 0.00 2.8 3.6 0.8 -1.58
ai_manip 10 25 47.64 34.03 45.0 47.19 44.48 1.0 99.0 98.0 0.17
condition 11 25 1.56 0.51 2.0 1.57 0.00 1.0 2.0 1.0 -0.23
kurtosis se
id -1.35 6.08
identity -1.44 6.47
consent 6.76 0.06
age -0.47 1.05
race 5.05 0.21
gender 12.17 0.14
manip_out -1.33 5.82
survey1 NaN 0.00
survey2 1.47 0.04
ai_manip -1.47 6.81
condition -2.02 0.10
------------------------------------------------------------
survey1: 2.6
vars n mean sd median trimmed mad min max range skew kurtosis se
id 1 1 44.0 NA 44.0 44.0 0 44.0 44.0 0 NA NA NA
identity 2 1 48.0 NA 48.0 48.0 0 48.0 48.0 0 NA NA NA
consent 3 1 2.0 NA 2.0 2.0 0 2.0 2.0 0 NA NA NA
age 4 1 28.0 NA 28.0 28.0 0 28.0 28.0 0 NA NA NA
race 5 1 6.0 NA 6.0 6.0 0 6.0 6.0 0 NA NA NA
gender 6 1 2.0 NA 2.0 2.0 0 2.0 2.0 0 NA NA NA
manip_out 7 1 30.0 NA 30.0 30.0 0 30.0 30.0 0 NA NA NA
survey1 8 1 2.6 NA 2.6 2.6 0 2.6 2.6 0 NA NA NA
survey2 9 1 3.4 NA 3.4 3.4 0 3.4 3.4 0 NA NA NA
ai_manip 10 1 52.0 NA 52.0 52.0 0 52.0 52.0 0 NA NA NA
condition 11 1 1.0 NA 1.0 1.0 0 1.0 1.0 0 NA NA NA
------------------------------------------------------------
survey1: 2.7
vars n mean sd median trimmed mad min max range skew kurtosis se
id 1 1 11.0 NA 11.0 11.0 0 11.0 11.0 0 NA NA NA
identity 2 1 55.0 NA 55.0 55.0 0 55.0 55.0 0 NA NA NA
consent 3 1 2.0 NA 2.0 2.0 0 2.0 2.0 0 NA NA NA
age 4 1 29.0 NA 29.0 29.0 0 29.0 29.0 0 NA NA NA
race 5 1 6.0 NA 6.0 6.0 0 6.0 6.0 0 NA NA NA
gender 6 1 2.0 NA 2.0 2.0 0 2.0 2.0 0 NA NA NA
manip_out 7 1 94.0 NA 94.0 94.0 0 94.0 94.0 0 NA NA NA
survey1 8 1 2.7 NA 2.7 2.7 0 2.7 2.7 0 NA NA NA
survey2 9 1 3.5 NA 3.5 3.5 0 3.5 3.5 0 NA NA NA
ai_manip 10 1 20.0 NA 20.0 20.0 0 20.0 20.0 0 NA NA NA
condition 11 1 1.0 NA 1.0 1.0 0 1.0 1.0 0 NA NA NA
------------------------------------------------------------
survey1: 3
vars n mean sd median trimmed mad min max range skew
id 1 13 54.00 31.87 67.0 54.73 29.65 5.0 95.0 90.0 -0.31
identity 2 13 46.69 30.41 34.0 46.00 28.17 9.0 92.0 83.0 0.31
consent 3 13 1.85 0.38 2.0 1.91 0.00 1.0 2.0 1.0 -1.70
age 4 13 31.23 9.72 28.0 29.45 1.48 24.0 58.0 34.0 1.66
race 5 13 4.00 1.91 3.0 4.00 1.48 1.0 7.0 6.0 0.20
gender 6 13 2.23 0.83 2.0 2.00 0.00 2.0 5.0 3.0 2.82
manip_out 7 13 53.15 30.86 55.0 52.73 38.55 11.0 100.0 89.0 -0.01
survey1 8 13 3.00 0.00 3.0 3.00 0.00 3.0 3.0 0.0 NaN
survey2 9 13 3.38 0.15 3.5 3.39 0.00 3.1 3.5 0.4 -0.54
ai_manip 10 13 63.38 26.60 67.0 65.00 10.38 9.0 100.0 91.0 -0.74
condition 11 13 1.62 0.51 2.0 1.64 0.00 1.0 2.0 1.0 -0.42
kurtosis se
id -1.68 8.84
identity -1.65 8.43
consent 0.99 0.10
age 1.68 2.70
race -1.58 0.53
gender 6.44 0.23
manip_out -1.62 8.56
survey1 NaN 0.00
survey2 -1.55 0.04
ai_manip -0.35 7.38
condition -1.96 0.14
------------------------------------------------------------
survey1: 3.09
vars n mean sd median trimmed mad min max range skew
id 1 2 61.00 35.36 61.00 61.00 37.06 36.00 86.00 50.0 0
identity 2 2 20.50 20.51 20.50 20.50 21.50 6.00 35.00 29.0 0
consent 3 2 2.00 0.00 2.00 2.00 0.00 2.00 2.00 0.0 NaN
age 4 2 26.00 2.83 26.00 26.00 2.97 24.00 28.00 4.0 0
race 5 2 3.50 0.71 3.50 3.50 0.74 3.00 4.00 1.0 0
gender 6 2 2.00 0.00 2.00 2.00 0.00 2.00 2.00 0.0 NaN
manip_out 7 2 33.50 40.31 33.50 33.50 42.25 5.00 62.00 57.0 0
survey1 8 2 3.09 0.00 3.09 3.09 0.00 3.09 3.09 0.0 NaN
survey2 9 2 3.25 0.35 3.25 3.25 0.37 3.00 3.50 0.5 0
ai_manip 10 2 48.50 51.62 48.50 48.50 54.11 12.00 85.00 73.0 0
condition 11 2 1.50 0.71 1.50 1.50 0.74 1.00 2.00 1.0 0
kurtosis se
id -2.75 25.00
identity -2.75 14.50
consent NaN 0.00
age -2.75 2.00
race -2.75 0.50
gender NaN 0.00
manip_out -2.75 28.50
survey1 NaN 0.00
survey2 -2.75 0.25
ai_manip -2.75 36.50
condition -2.75 0.50
------------------------------------------------------------
survey1: 3.1
vars n mean sd median trimmed mad min max range skew kurtosis
id 1 5 52.2 25.82 47.0 52.2 19.27 29.0 94.0 65 0.63 -1.45
identity 2 5 58.4 26.82 66.0 58.4 13.34 12.0 78.0 66 -0.91 -1.12
consent 3 5 1.8 0.45 2.0 1.8 0.00 1.0 2.0 1 -1.07 -0.92
age 4 5 30.0 8.03 27.0 30.0 2.97 24.0 44.0 20 0.94 -1.07
race 5 5 4.4 1.52 4.0 4.4 1.48 3.0 6.0 3 0.15 -2.21
gender 6 5 2.0 0.00 2.0 2.0 0.00 2.0 2.0 0 NaN NaN
manip_out 7 5 57.2 26.27 45.0 57.2 14.83 35.0 95.0 60 0.42 -1.93
survey1 8 5 3.1 0.00 3.1 3.1 0.00 3.1 3.1 0 NaN NaN
survey2 9 5 3.2 0.42 3.4 3.2 0.15 2.5 3.5 1 -0.74 -1.41
ai_manip 10 5 65.2 34.67 81.0 65.2 20.76 22.0 95.0 73 -0.27 -2.17
condition 11 5 1.4 0.55 1.0 1.4 0.00 1.0 2.0 1 0.29 -2.25
se
id 11.55
identity 11.99
consent 0.20
age 3.59
race 0.68
gender 0.00
manip_out 11.75
survey1 0.00
survey2 0.19
ai_manip 15.50
condition 0.24
------------------------------------------------------------
survey1: 3.18
vars n mean sd median trimmed mad min max range skew
id 1 4 44.25 32.39 42.00 44.25 37.06 13.00 80.00 67.0 0.07
identity 2 4 27.25 19.38 19.50 27.25 5.19 14.00 56.00 42.0 0.70
consent 3 4 2.00 0.00 2.00 2.00 0.00 2.00 2.00 0.0 NaN
age 4 4 28.25 1.26 28.00 28.25 0.74 27.00 30.00 3.0 0.42
race 5 4 3.50 1.73 3.00 3.50 0.74 2.00 6.00 4.0 0.58
gender 6 4 2.75 1.50 2.00 2.75 0.00 2.00 5.00 3.0 0.75
manip_out 7 4 36.50 35.82 28.00 36.50 21.50 3.00 87.00 84.0 0.48
survey1 8 4 3.18 0.00 3.18 3.18 0.00 3.18 3.18 0.0 NaN
survey2 9 4 3.52 0.05 3.50 3.52 0.00 3.50 3.60 0.1 0.75
ai_manip 10 4 53.25 39.14 60.00 53.25 37.06 5.00 88.00 83.0 -0.21
condition 11 4 1.50 0.58 1.50 1.50 0.74 1.00 2.00 1.0 0.00
kurtosis se
id -2.32 16.19
identity -1.72 9.69
consent NaN 0.00
age -1.82 0.63
race -1.77 0.87
gender -1.69 0.75
manip_out -1.82 17.91
survey1 NaN 0.00
survey2 -1.69 0.03
ai_manip -2.19 19.57
condition -2.44 0.29
------------------------------------------------------------
survey1: 3.2
vars n mean sd median trimmed mad min max range skew kurtosis
id 1 2 45.00 35.36 45.00 45.00 37.06 20.0 70.0 50.0 0 -2.75
identity 2 2 69.00 22.63 69.00 69.00 23.72 53.0 85.0 32.0 0 -2.75
consent 3 2 2.00 0.00 2.00 2.00 0.00 2.0 2.0 0.0 NaN NaN
age 4 2 47.00 25.46 47.00 47.00 26.69 29.0 65.0 36.0 0 -2.75
race 5 2 5.00 2.83 5.00 5.00 2.97 3.0 7.0 4.0 0 -2.75
gender 6 2 2.00 0.00 2.00 2.00 0.00 2.0 2.0 0.0 NaN NaN
manip_out 7 2 63.00 16.97 63.00 63.00 17.79 51.0 75.0 24.0 0 -2.75
survey1 8 2 3.20 0.00 3.20 3.20 0.00 3.2 3.2 0.0 NaN NaN
survey2 9 2 3.45 0.07 3.45 3.45 0.07 3.4 3.5 0.1 0 -2.75
ai_manip 10 2 36.00 9.90 36.00 36.00 10.38 29.0 43.0 14.0 0 -2.75
condition 11 2 1.50 0.71 1.50 1.50 0.74 1.0 2.0 1.0 0 -2.75
se
id 25.00
identity 16.00
consent 0.00
age 18.00
race 2.00
gender 0.00
manip_out 12.00
survey1 0.00
survey2 0.05
ai_manip 7.00
condition 0.50
------------------------------------------------------------
survey1: 3.27
vars n mean sd median trimmed mad min max range skew
id 1 3 69.00 25.36 58.00 69.00 10.38 51.00 98.00 47.0 0.35
identity 2 3 23.67 21.73 20.00 23.67 23.72 4.00 47.00 43.0 0.16
consent 3 3 2.00 0.00 2.00 2.00 0.00 2.00 2.00 0.0 NaN
age 4 3 26.00 3.46 28.00 26.00 0.00 22.00 28.00 6.0 -0.38
race 5 3 4.33 1.53 4.00 4.33 1.48 3.00 6.00 3.0 0.21
gender 6 3 2.00 0.00 2.00 2.00 0.00 2.00 2.00 0.0 NaN
manip_out 7 3 68.33 15.37 61.00 68.33 4.45 58.00 86.00 28.0 0.37
survey1 8 3 3.27 0.00 3.27 3.27 0.00 3.27 3.27 0.0 NaN
survey2 9 3 3.33 0.29 3.50 3.33 0.00 3.00 3.50 0.5 -0.38
ai_manip 10 3 53.00 26.66 44.00 53.00 17.79 32.00 83.00 51.0 0.30
condition 11 3 2.00 0.00 2.00 2.00 0.00 2.00 2.00 0.0 NaN
kurtosis se
id -2.33 14.64
identity -2.33 12.55
consent NaN 0.00
age -2.33 2.00
race -2.33 0.88
gender NaN 0.00
manip_out -2.33 8.88
survey1 NaN 0.00
survey2 -2.33 0.17
ai_manip -2.33 15.39
condition NaN 0.00
------------------------------------------------------------
survey1: 3.5
vars n mean sd median trimmed mad min max range skew
id 1 40 48.42 27.78 45.5 47.72 29.65 3.0 100.0 97.0 0.22
identity 2 40 55.77 26.20 54.5 56.31 27.43 2.0 100.0 98.0 -0.08
consent 3 40 1.98 0.16 2.0 2.00 0.00 1.0 2.0 1.0 -5.86
age 4 40 27.82 5.34 26.0 26.84 2.97 21.0 48.0 27.0 1.97
race 5 40 4.15 1.53 3.5 4.19 0.74 2.0 6.0 4.0 0.22
gender 6 40 2.25 1.03 2.0 2.00 0.00 1.0 7.0 6.0 3.33
manip_out 7 40 50.30 30.47 46.0 49.91 36.32 2.0 99.0 97.0 0.08
survey1 8 40 3.50 0.00 3.5 3.50 0.00 3.5 3.5 0.0 NaN
survey2 9 40 3.39 0.16 3.5 3.42 0.00 3.0 3.5 0.5 -1.33
ai_manip 10 40 47.52 26.67 47.5 47.47 35.58 3.0 93.0 90.0 0.03
condition 11 40 1.40 0.50 1.0 1.38 0.00 1.0 2.0 1.0 0.39
kurtosis se
id -1.13 4.39
identity -1.03 4.14
consent 33.15 0.02
age 4.28 0.84
race -1.68 0.24
gender 10.82 0.16
manip_out -1.40 4.82
survey1 NaN 0.00
survey2 0.44 0.03
ai_manip -1.33 4.22
condition -1.89 0.08
------------------------------------------------------------
survey1: 3.6
vars n mean sd median trimmed mad min max range skew kurtosis se
id 1 1 90.0 NA 90.0 90.0 0 90.0 90.0 0 NA NA NA
identity 2 1 19.0 NA 19.0 19.0 0 19.0 19.0 0 NA NA NA
consent 3 1 1.0 NA 1.0 1.0 0 1.0 1.0 0 NA NA NA
age 4 1 28.0 NA 28.0 28.0 0 28.0 28.0 0 NA NA NA
race 5 1 3.0 NA 3.0 3.0 0 3.0 3.0 0 NA NA NA
gender 6 1 2.0 NA 2.0 2.0 0 2.0 2.0 0 NA NA NA
manip_out 7 1 36.0 NA 36.0 36.0 0 36.0 36.0 0 NA NA NA
survey1 8 1 3.6 NA 3.6 3.6 0 3.6 3.6 0 NA NA NA
survey2 9 1 3.5 NA 3.5 3.5 0 3.5 3.5 0 NA NA NA
ai_manip 10 1 50.0 NA 50.0 50.0 0 50.0 50.0 0 NA NA NA
condition 11 1 2.0 NA 2.0 2.0 0 2.0 2.0 0 NA NA NA
------------------------------------------------------------
survey1: 3.63
vars n mean sd median trimmed mad min max range skew kurtosis se
id 1 1 93.00 NA 93.00 93.00 0 93.00 93.00 0 NA NA NA
identity 2 1 7.00 NA 7.00 7.00 0 7.00 7.00 0 NA NA NA
consent 3 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA
age 4 1 24.00 NA 24.00 24.00 0 24.00 24.00 0 NA NA NA
race 5 1 5.00 NA 5.00 5.00 0 5.00 5.00 0 NA NA NA
gender 6 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA
manip_out 7 1 57.00 NA 57.00 57.00 0 57.00 57.00 0 NA NA NA
survey1 8 1 3.63 NA 3.63 3.63 0 3.63 3.63 0 NA NA NA
survey2 9 1 3.00 NA 3.00 3.00 0 3.00 3.00 0 NA NA NA
ai_manip 10 1 30.00 NA 30.00 30.00 0 30.00 30.00 0 NA NA NA
condition 11 1 2.00 NA 2.00 2.00 0 2.00 2.00 0 NA NA NA#
# # also use histograms and scatterplots to examine your continuous variables
hist(d$survey2)
plot(d$survey1, d$survey2)
#
# # and table() and cross_cases() to examine your categorical variables
# # you may not need the cross_cases code
# table(d$IV)
# cross_cases(d, IV1, IV2)
#
# # and boxplot to examine any categorical variables with continuous variables
boxplot(d$survey2~d$condition)
#
# #convert any categorical variables to factors
# d$variable <- as.factor(d$variable)Check Your Assumptions
t-Test Assumptions
- Data values must be independent (independent t-test only) (confirmed by data report)
- Data obtained via a random sample (confirmed by data report)
- IV must have two levels (will check below)
- Dependent variable must be normally distributed (will check below. if issues, note and proceed)
- Variances of the two groups must be approximately equal, aka ‘homogeneity of variance’. Lacking this makes our results inaccurate (will check below - this really only applies to Student’s t-test, but we’ll check it anyway)
Checking IV levels
# # preview the levels and counts for your IV
table(d$condition, useNA = "always")
1 2 <NA>
50 50 0 #
# # note that the table() output shows you exactly how the levels of your variable are written. when recoding, make sure you are spelling them exactly as they appear
#
# # to drop levels from your variable
# # this subsets the data and says that any participant who is coded as 'BAD' should be removed
# d <- subset(d, IV != "BAD")
#
table(d$condition, useNA = "always")
1 2 <NA>
50 50 0 #
# # to combine levels
# # this says that where any participant is coded as 'BAD' it should be replaced by 'GOOD'
# d$iv_rc[d$iv == "BAD"] <- "GOOD"
#
table(d$condition, useNA = "always")
1 2 <NA>
50 50 0 #
# # check your variable types
str(d)'data.frame': 100 obs. of 11 variables:
$ id : int 1 2 3 4 5 6 7 8 9 10 ...
$ identity : chr "I’m a 29-year-old White woman from Ohio. I’m an introvert, often battling anxiety and feelings of loneliness. T"| __truncated__ "I'm a 27-year-old Black woman living in Atlanta, juggling my job as a graphic designer with freelance gigs. I s"| __truncated__ "I’m a 29-year-old Asian American woman, navigating the complexities of my career in advertising. I often feel p"| __truncated__ "I’m a 28-year-old White woman living in Portland. I’m passionate about sustainable living but often feel overwh"| __truncated__ ...
$ consent : chr "I understand these instructions." "I understand these instructions." "I understand these instructions." "I understand these instructions." ...
$ age : int 29 27 29 28 27 32 32 24 24 29 ...
$ race : int 6 3 2 6 3 6 6 6 4 7 ...
$ gender : int 2 2 2 2 2 2 2 2 2 2 ...
$ manip_out: chr "Thank you for sharing that context about yourself. It sounds like you're navigating a complex relationship with"| __truncated__ "Thank you for sharing this information. It sounds like you're participating in an interesting study that could "| __truncated__ "Thank you for sharing that context; it sounds like you're navigating a challenging but important journey. Parti"| __truncated__ "Thank you for the context! I'm here to help you explore the themes related to your identity and how they inters"| __truncated__ ...
$ survey1 : num 2.27 2.5 3.5 2.5 3 2.5 3.5 2.5 3.5 3 ...
$ survey2 : num 3.5 3.5 3.5 3.5 3.2 3 3.5 3.5 3.2 3.1 ...
$ ai_manip : chr "Thank you for sharing more about your background and experiences. It sounds like you're reflecting thoughtfully"| __truncated__ "It sounds like you're navigating some common challenges that many people face, especially those in creative fie"| __truncated__ "It sounds like you're reflecting deeply on your experiences and feelings related to social media, your career, "| __truncated__ "It sounds like you're looking for a deeper understanding of how your identity, experiences, and personality tra"| __truncated__ ...
$ condition: int 1 1 1 1 1 1 1 1 1 1 ...#
# # make sure that your IV is recognized as a factor by R
# # if you created a new _rc variable make sure to use that one instead
d$condition <- as.factor(d$condition)Testing Homogeneity of Variance with Levene’s Test
We can test whether the variances of our two groups are equal using Levene’s test. The null hypothesis is that the variance between the two groups is equal, which is the result we want. So when running Levene’s test we’re hoping for a non-significant result!
# # use the leveneTest() command from the car package to test homogeneity of variance
# # uses the same 'formula' setup that we'll use for our t-test: formula is y~x, where y is our DV and x is our IV
leveneTest(survey2~condition, data = d)Levene's Test for Homogeneity of Variance (center = median)
Df F value Pr(>F)
group 1 0.4529 0.5026
98 Pearson’s Correlation Coefficient Assumptions
- Should have two measurements for each participant for each variable (confirmed by earlier procedures – we dropped any participants with missing data)
- Variables should be continuous and normally distributed, or assessments of the relationship may be inaccurate (will do below)
- Outliers should be identified and removed, or results will be inaccurate (will do below)
- Relationship between the variables should be linear, or they will not be detected (will do below)
Run a Multiple Linear Regression
To check the assumptions for Pearson’s correlation coefficient, we run our regression and then check our diagnostic plots.
# # use the lm() command to run the regression
# # dependent/outcome variable on the left, independent/predictor variables on the right
reg_model <- lm(survey2 ~ survey1, data = d)Check linearity with Residuals vs Fitted plot
For some examples of good Residuals vs Fitted plot and ones that show serious errors, check out this page.
For your homework, you’ll simply need to generate this plot and talk about how your plot compares to the good and problematic plots linked to above. Is it closer to the ‘good’ plots or one of the ‘bad’ plots? This is going to be a judgement call, and that’s okay! In practice, you’ll always be making these judgement calls as part of a team, so this assignment is just about getting experience with it, not making the perfect call.
plot(reg_model, 1)
Check for outliers using Cook’s distance and a Residuals vs Leverage plot
For your homework, you’ll simply need to generate these plots, assess Cook’s distance in your dataset, and then identify any potential cases that are prominent outliers.
# # Cook's distance
plot(reg_model, 4)
#
# # Residuals vs Leverage
plot(reg_model, 5)
Issues with My Data
Describe any issues and why they’re problematic here.
Run Your Analysis
Run a t-Test
# # very simple! we specify the dataframe alongside the variables instead of having a separate argument for the dataframe like we did for leveneTest()
t_output <- t.test(d$survey2~d$condition)View Test Output
t_output
Welch Two Sample t-test
data: d$survey2 by d$condition
t = -0.87038, df = 90.974, p-value = 0.3864
alternative hypothesis: true difference in means between group 1 and group 2 is not equal to 0
95 percent confidence interval:
-0.1115952 0.0435952
sample estimates:
mean in group 1 mean in group 2
3.368 3.402 Calculate Cohen’s d
# # once again, we use our formula to calculate cohen's d
d_output <- cohen.d(d$survey2~d$condition)View Effect Size
- Trivial: < .2
- Small: between .2 and .5
- Medium: between .5 and .8
- Large: > .8
d_output
Cohen's d
d estimate: -0.1740754 (negligible)
95 percent confidence interval:
lower upper
-0.5717199 0.2235690 Run a Correlation Test
Create a Correlation Matrix
d2 <- subset(d, select= c(survey1, survey2))
corr_output_m <- corr.test(d2)View Test Output
- Strong effect: Between |0.50| and |1|
- Moderate effect: Between |0.30| and |0.49|
- Weak effect: Between |0.10| and |0.29|
- Trivial effect: Less than |0.09|
corr_output_mCall:corr.test(x = d2)
Correlation matrix
survey1 survey2
survey1 1.00 -0.06
survey2 -0.06 1.00
Sample Size
[1] 100
Probability values (Entries above the diagonal are adjusted for multiple tests.)
survey1 survey2
survey1 0.00 0.58
survey2 0.58 0.00
To see confidence intervals of the correlations, print with the short=FALSE optionWrite Up Results
t-Test
he hypothesis predicted that participants in the social media use condition would have higher post-stress scores than those in the control condition. The independent variable was condition (social media use vs. control), and the dependent variable was post-survey stress scores (survey2).
Prior to conducting the t-test, I ran Levene’s Test for homogeneity of variance to check if the variances between the two groups were equal. Levene’s test was not significant, 𝐹 ( 1 , 98 ) = 0.45 , 𝑝 = .503 F(1,98)=0.45,p=.503, indicating that the assumption of equal variances was met.
An independent samples t-test was then performed to compare post-stress scores between the two conditions. The results indicated no significant difference between the social media use and control conditions, t(90.97)=−0.87,p=.386. The means for the social media use condition (M = 3.37, SD = 0.89) and the control condition (M = 3.40, SD = 0.91) were very similar, with a 95% confidence interval for the difference in means ranging from -0.11 to 0.04. Cohen’s d was calculated as -0.17, which indicates a negligible effect size.
There were no major issues, and the analysis was conducted successfully. However, it is important to note that the assumption of equal variances was confirmed, and the results of the t-test were consistent with that assumption.
Correlation Test
The hypothesis predicted that pre-survey stress scores (survey1) would be positively correlated with post-survey stress scores (survey2) across all participants. To test this, a Pearson correlation was conducted between the pre- and post-survey stress scores.
The correlation between pre-survey and post-survey stress scores was weak and negative, r = -0.06, p = .58, suggesting no meaningful relationship between the two variables. The 95% confidence interval for the correlation ranged from -0.18 to 0.09. This result was not statistically significant, indicating that pre-survey stress scores did not reliably predict post-survey stress scores.
There were no issues with the coding, and the analysis was conducted successfully. However, the weak, non-significant correlation suggests that the relationship between pre- and post-survey stress scores is minimal.
References
Cohen J. (1988). Statistical Power Analysis for the Behavioral Sciences. New York, NY: Routledge Academic.