Learning Log 8
Jade Gurtala
24/07/2021
This week's coding goals
This week, my goal was to start thinking of questions for my exploratory analyses and code the descriptive statistics, figure and statistics for at least one question.
The Questions
I brainstormed a couple of questions, like:
- Is there an effect of gender on changes in implicit biases?
- Or an alternate question: Are males more biased than women?
- Is the number of cues that participants are exposed to influence changes in implicit bias?
- (Number of cues range from 37 to 660)
- Does the difference in procedure (between the original study and the replication study) matter - does getting offered course credit (vs. cash) significantly change implicit biases?
- Are the number of years spent in university affect average bias scores for each IAT timepoint, averaged over the cued and uncued conditions?
I'm still not sure which 3 questions I'll choose, but I will definitely choose the last question.
So I started with that question:
- Do the number of years spent in university affect students' bias scores?
- I've defined this question as the average bias scores recorded at each of the 4 IAT timepoints, averaged over the cued and uncued conditions.
- My graph will look like a line graph with 4 different coloured lines, connoting the number of years spent in university (0, 1, 2, or 3). The x-axis will have the 4 IAT timepoints (baseline, prenap, postnap, and one-week delay), and the y-axis will have their average bias scores.
How did I go?
Preliminaries
First, I loaded the packages and read in the data.
load packages
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──## ✓ ggplot2 3.3.4 ✓ purrr 0.3.4
## ✓ tibble 3.1.2 ✓ dplyr 1.0.7
## ✓ tidyr 1.1.3 ✓ stringr 1.4.0
## ✓ readr 1.4.0 ✓ forcats 0.5.1## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(readspss)
library(ggplot2)
library(janitor)##
## Attaching package: 'janitor'## The following objects are masked from 'package:stats':
##
## chisq.test, fisher.testlibrary(plotrix)
library(gt)Read in the data
cleandata <- read_csv("cleandata.csv")##
## ── Column specification ────────────────────────────────────────────────────────
## cols(
## .default = col_character(),
## General_1_Age = col_double(),
## General_1_EnglishYrs = col_double(),
## General_1_CaffCups = col_double(),
## General_1_CaffHrsAgo = col_double(),
## General_1_UniYears = col_double(),
## Epworth_total = col_double(),
## AlertTest_1_Concentr_1 = col_double(),
## AlertTest_1_Refresh_1 = col_double(),
## AlertTest_2_Concentr_1 = col_double(),
## AlertTest_2_Refresh_1 = col_double(),
## AlertTest_3_Concentr_1 = col_double(),
## AlertTest_3_Refresh_1 = col_double(),
## AlertTest_4_Concentr_1 = col_double(),
## AlertTest_4_Refresh_1 = col_double(),
## Total_sleep = col_double(),
## Wake_amount = col_double(),
## NREM1_amount = col_double(),
## NREM2_amount = col_double(),
## SWS_amount = col_double(),
## REM_amount = col_double()
## # ... with 26 more columns
## )
## ℹ Use `spec()` for the full column specifications.Descriptive statistics
First, I have to calculate the mean and standard errors for my plot later on.
Attempt 1
- I calculated the average bias scores by first selecting my needed variables from my data file using
select(). Then, usingsummarise()to list the means for each variable containing "IAT" usingacross()andcontains(). -head()is used to view the calculated data
avgbias <- cleandata %>%
select(baseIATcued, baseIATuncued, preIATcued, preIATuncued, postIATcued, postIATuncued,weekIATcued, weekIATuncued, General_1_UniYears)
avgbias <- avgbias %>%
summarise(across(contains("IAT"), list(mean = mean)))
head(avgbias)## # A tibble: 1 x 8
## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.518 0.595 0.211 0.302
## # … with 4 more variables: postIATcued_mean <dbl>, postIATuncued_mean <dbl>,
## # weekIATcued_mean <dbl>, weekIATuncued_mean <dbl>- To calculate standard error, I used
std.error()to calculate from the means I calculated above. - However, it kept coming up with NA for my standard errors.
se_bias <- std.error(avgbias)
head(se_bias)## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## NA NA NA NA
## postIATcued_mean postIATuncued_mean
## NA NAAttempt 2
I tried again, but now I decided to calculate means and standard errors for each of the 4 conditions (number of years spent at Uni).
0 years at Uni bias change:
First, I selected participants that have only attended uni for 0 years, using filter()
year0uni_participants <- cleandata %>%
filter(General_1_UniYears == "0")I calculated average bias levels and standard error for this group, using the same method as earlier:
year0uni <- year0uni_participants %>%
select(baseIATcued, baseIATuncued, preIATcued, preIATuncued, postIATcued, postIATuncued,weekIATcued, weekIATuncued)
year0uni_avgbias <- year0uni %>%
summarise(across(contains("IAT"), list(mean = mean)))
print(year0uni_avgbias)## # A tibble: 1 x 8
## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.672 0.455 0.408 0.269
## # … with 4 more variables: postIATcued_mean <dbl>, postIATuncued_mean <dbl>,
## # weekIATcued_mean <dbl>, weekIATuncued_mean <dbl>year0uni_se <- std.error(year0uni_avgbias)
head(year0uni_se)## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## NA NA NA NA
## postIATcued_mean postIATuncued_mean
## NA NA- However, for standard error, it kept coming up with the same NA error. Thus, I tried using
na.rmto get rid of the NAs.
year0uni_se <- std.error(year0uni_avgbias, na.rm = TRUE)
head(year0uni_se)## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## NA NA NA NA
## postIATcued_mean postIATuncued_mean
## NA NAThen, I realised I was using the wrong mean variable to calculate standard error. Now that I have calculated using the right variable, it has calculated for standard error.
year0uni_se <- std.error(year0uni)
head(year0uni_se)## baseIATcued baseIATuncued preIATcued preIATuncued postIATcued
## 0.1124953 0.1401144 0.1231403 0.1006117 0.1505249
## postIATuncued
## 0.1604736Now I have to do the same for Years 1,2,and 3: >Year 1
First, I selected participants that have only attended uni for 1 year, using filter()
year1uni_participants <- cleandata %>%
filter(General_1_UniYears == "1")Then, calculated avg bias levels for that group:
year1uni <- year1uni_participants %>%
select(baseIATcued, baseIATuncued, preIATcued, preIATuncued, postIATcued, postIATuncued,weekIATcued, weekIATuncued)
year1uni_avgbias <- year1uni %>%
summarise(across(contains("IAT"), list(mean = mean)))
head(year1uni_avgbias)## # A tibble: 1 x 8
## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.509 0.742 0.305 0.390
## # … with 4 more variables: postIATcued_mean <dbl>, postIATuncued_mean <dbl>,
## # weekIATcued_mean <dbl>, weekIATuncued_mean <dbl>Then, calculated standard error using the means above:
year1uni_se <- std.error(year1uni)
head(year1uni_se)## baseIATcued baseIATuncued preIATcued preIATuncued postIATcued
## 0.2331138 0.2359056 0.3046492 0.2841019 0.1672415
## postIATuncued
## 0.2697327Year 2
First, I selected participants that have only attended uni for 2 years, using filter()
year2uni_participants <- cleandata %>%
filter(General_1_UniYears == "2")Then, calculated avg bias levels for that group:
year2uni <- year2uni_participants %>%
select(baseIATcued, baseIATuncued, preIATcued, preIATuncued, postIATcued, postIATuncued,weekIATcued, weekIATuncued)
year2uni_avgbias <- year2uni %>%
summarise(across(contains("IAT"), list(mean = mean)))
head(year2uni_avgbias)## # A tibble: 1 x 8
## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.340 0.719 -0.0141 0.371
## # … with 4 more variables: postIATcued_mean <dbl>, postIATuncued_mean <dbl>,
## # weekIATcued_mean <dbl>, weekIATuncued_mean <dbl>Then, calculated standard error using the means above:
year2uni_se <- std.error(year2uni)
head(year2uni_se)## baseIATcued baseIATuncued preIATcued preIATuncued postIATcued
## 0.09853969 0.15985863 0.20114450 0.13174385 0.13812838
## postIATuncued
## 0.14299955Year 3
First, I selected participants that have only attended uni for 3 years, using filter()
year3uni_participants <- cleandata %>%
filter(General_1_UniYears == "3")Then, calculated avg bias levels for that group:
year3uni <- year3uni_participants %>%
select(baseIATcued, baseIATuncued, preIATcued, preIATuncued, postIATcued, postIATuncued,weekIATcued, weekIATuncued)
year3uni_avgbias <- year1uni %>%
summarise(across(contains("IAT"), list(mean = mean)))
head(year3uni_avgbias)## # A tibble: 1 x 8
## baseIATcued_mean baseIATuncued_mean preIATcued_mean preIATuncued_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.509 0.742 0.305 0.390
## # … with 4 more variables: postIATcued_mean <dbl>, postIATuncued_mean <dbl>,
## # weekIATcued_mean <dbl>, weekIATuncued_mean <dbl>Then, calculated standard error using the means above:
year3uni_se <- std.error(year3uni)
head(year3uni_se)## baseIATcued baseIATuncued preIATcued preIATuncued postIATcued
## 0.1240071 0.1158314 0.1314291 0.2613532 0.1547907
## postIATuncued
## 0.1680495Create the data tibble
Because I want to see if there are bias changes over time, for the number of years, I used a line graph:
biasdata <- data.frame(
condition = factor(c("0", "0", "0", "0", "1", "1", "1", "1", "2", "2", "2", "2", "3", "3", "3", "3")),
time = factor(c("Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week")),
levels = c("Baseline", "Prenap", "Postnap", "1-week"),
bias_av = c(year0uni$ ## Error: <text>:6:0: unexpected end of input
## 4: levels = c("Baseline", "Prenap", "Postnap", "1-week"),
## 5: bias_av = c(year0uni$
## ^Halfway through constructing the tibble, I realised that I need to create a column/variable that averages each IAT timepoint over the cued and uncued condition i.e. need to find IAT score for each timepoint averaged over cued/uncued condition. Use mutate?
Attempt 3
I have to start again from the beginning:
exploratorydata1 <- read.csv("cleandata.csv")Create baseline IAT, averaged over cued/uncued condition using mutate():
mutated_exploratorydata1 <- exploratorydata1 %>%
rowwise() %>%
mutate(baselineIAT = mean(c(baseIATcued, baseIATuncued)),
prenapIAT = mean(c(preIATcued, preIATuncued)),
postnapIAT = mean(c(postIATcued, postIATuncued)),
weekIAT = mean(c(weekIATcued, weekIATuncued)))Used glimpse() to view if the data calculated properly - it did as there are now 105 columns, rather than 101 columns.
glimpse(mutated_exploratorydata1)## Rows: 31
## Columns: 105
## Rowwise:
## $ ParticipantID <chr> "ub6", "ub7", "ub8", "ub9", "ub11", "ub13", "ub…
## $ exclude <chr> "no", "no", "no", "no", "no", "no", "no", "no",…
## $ cue_presented <chr> "yes", "yes", "yes", "yes", "yes", "yes", "yes"…
## $ heard_cue_report <chr> "no", "no", "no", "no", "no", "no", "no", "no",…
## $ heard_cue_exit <chr> "no", "unsure", "no", "no", "no", "no", "no", "…
## $ predicted_cue <chr> "no", "no", "no", "suspected", "no", "no", "no"…
## $ Cue_condition <chr> "race cue played", "race cue played", "gender c…
## $ Counterbias_order <chr> "racial training first", "gender training first…
## $ Sound_assignment <chr> "machR and descG", "machG and descR", "machR an…
## $ IAT1_order <chr> "EATF-SATF", "SATF-EATF", "EATF-SATF", "SATF-EA…
## $ IAT234_order <chr> "SATS-EATS", "EATS-SATS", "SATS-EATS", "EATS-SA…
## $ IAT_order <chr> "ES, SESESE", "SE, ESESES", "ES, SESESE", "SE, …
## $ compensation <chr> "cash", "cash", "cash", "cash", "course credit"…
## $ General_1_Age <int> 21, 21, 20, 21, 19, 20, 18, 18, 18, 18, 19, 19,…
## $ General_1_Sex <chr> "Female", "Female", "Female", "Male", "Female",…
## $ General_1_Race <chr> "White", "White", "White", "White", "White", "W…
## $ General_1_English <chr> "Yes", "Yes", "Yes", "Yes", "No", "Yes", "Yes",…
## $ General_1_EnglishYrs <int> NA, NA, NA, NA, 12, NA, NA, NA, NA, NA, NA, NA,…
## $ General_1_Caffeine <chr> "Yes", "Yes", "No", "No", "No", "No", "No", "Ye…
## $ General_1_CaffCups <int> 1, 1, NA, NA, NA, NA, NA, 1, NA, 1, NA, NA, 1, …
## $ General_1_CaffHrsAgo <dbl> 2.0, 3.0, NA, NA, NA, NA, NA, 5.5, NA, 2.0, NA,…
## $ General_1_SleepDisor <chr> "No", "No", "No", "Yes", "No", "No", "No", "No"…
## $ General_1_MentalDiso <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ General_1_Meds <chr> "No", "Yes", "No", "No", "Yes", "Yes", "No", "N…
## $ General_1_MedList <chr> "", "20 mg prozac every day", "", "", "birth co…
## $ General_1_University <chr> "Furman University", "Furman University", "Furm…
## $ General_1_UniYears <int> 3, 3, 2, 3, 0, 2, 0, 0, 0, 0, 1, 0, 0, 0, 1, 2,…
## $ Demo_1_Ethnic <chr> "Not Hispanic or Latino", "Not Hispanic or Lati…
## $ Demo_1_Racial <chr> "White", "White", "White", "White", "White", "W…
## $ Demo_1_Gender <chr> "Female", "Female", "Female", "Male", "Female",…
## $ Demo_1_NonParticipat <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
## $ Epworth_1_Read <chr> "slight chance of dozing", "slight chance of do…
## $ Epworth_1_TV <chr> "slight chance of dozing", "slight chance of do…
## $ Epworth_1_Public <chr> "slight chance of dozing", "no chance of dozing…
## $ Epworth_1_Passenger <chr> "moderate chance of dozing", "slight chance of …
## $ Epworth_1_LyingDown <chr> "high chance of dozing", "slight chance of dozi…
## $ Epworth_1_Talking <chr> "no chance of dozing", "no chance of dozing", "…
## $ Epworth_1_Lunch <chr> "no chance of dozing", "no chance of dozing", "…
## $ Epworth_1_Traffic <chr> "no chance of dozing", "no chance of dozing", "…
## $ Epworth_total <int> 16, 12, 13, 10, 16, 12, 20, 15, 16, 16, 11, 21,…
## $ AlertTest_1_Concentr_1 <int> 80, 60, 60, 70, 70, 40, 80, 80, 80, 50, 90, 30,…
## $ AlertTest_1_Refresh_1 <int> 60, 70, 60, 30, 60, 40, 80, 60, 40, 30, 90, 40,…
## $ AlertTest_1_Feel <chr> "2 - Functioning at high levels, but not at pea…
## $ AlertTest_2_Concentr_1 <int> 70, 60, 40, 60, 80, 40, 80, 80, 60, 40, 90, 20,…
## $ AlertTest_2_Refresh_1 <int> 70, 60, 30, 30, 60, 40, 70, 60, 40, 30, 70, 30,…
## $ AlertTest_2_Feel <chr> "3 - Awake, but relaxed; responsive but not ful…
## $ AlertTest_3_Concentr_1 <int> NA, 60, 40, 80, NA, 60, 100, 80, 70, NA, 70, 50…
## $ AlertTest_3_Refresh_1 <int> NA, 70, 50, 70, NA, 80, 100, 90, 90, NA, 100, 7…
## $ AlertTest_3_Feel <chr> NA, "2 - Functioning at high levels, but not at…
## $ AlertTest_4_Concentr_1 <int> 80, 60, 40, NA, NA, 70, 90, 100, 70, 80, 100, 3…
## $ AlertTest_4_Refresh_1 <int> 90, 50, 30, NA, NA, 80, 90, 80, 60, 80, 100, 70…
## $ AlertTest_4_Feel <chr> "1 - Feeling active, vital alert, or wide awake…
## $ S1_ExitQ_1_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S1_ExitQ_1_soundaffect <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S1_ExitQ_2_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S1_ExitQ_3_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S1_ExitQ_4_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S1_ExitQ_4_soundaffect <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S1_ExitQ_5_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S1_ExitQ_5_soundaffect <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_1_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_1_soundaffect <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_2_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_3_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_4_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_4_soundaffect <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_5_sound <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ S2_ExitQ_5_soundaffect <chr> "No", "No", "No", "No", "No", "No", "No", "No",…
## $ Total_sleep <int> 65, 66, 80, 62, 51, 81, 81, 67, 79, 71, 68, 85,…
## $ Wake_amount <int> 25, 24, 10, 28, 39, 9, 9, 23, 11, 19, 22, 5, 16…
## $ NREM1_amount <int> 10, 9, 5, 5, 11, 4, 3, 4, 4, 2, 15, 4, 4, 7, 3,…
## $ NREM2_amount <dbl> 20.0, 52.0, 15.0, 15.5, 22.0, 23.0, 31.0, 50.0,…
## $ SWS_amount <int> 12, 19, 24, 24, 16, 36, 37, 4, 23, 25, 7, 31, 4…
## $ REM_amount <int> 23, 0, 17, 17, 2, 18, 9, 9, 24, 11, 10, 16, 0, …
## $ SWSxREM <int> 276, 0, 408, 408, 32, 648, 333, 36, 552, 275, 7…
## $ cue_minutes <dbl> 9.5, 12.0, 15.5, 16.0, 15.0, 25.0, 23.0, 3.0, 1…
## $ baseIATcued <dbl> 0.57544182, 0.09911241, 0.20577365, 0.35314196,…
## $ baseIATuncued <dbl> 0.60953653, 0.64396538, 1.52435622, 0.13108478,…
## $ preIATcued <dbl> 0.55905291, -0.13380639, 0.51077026, -0.0293319…
## $ preIATuncued <dbl> 0.21462144, 0.33985028, 0.37990232, -0.94209553…
## $ postIATcued <dbl> 0.681910146, 0.044634805, -0.002583615, -0.2459…
## $ postIATuncued <dbl> 0.46728694, -0.05686262, 0.68243589, 0.94970369…
## $ weekIATcued <dbl> 0.20377367, 0.45873715, 0.39859469, 0.92341592,…
## $ weekIATuncued <dbl> 0.68277422, -0.01070460, 0.71187286, 0.20212832…
## $ postnap_change_cued <dbl> 0.1228572, 0.1784412, -0.5133539, -0.2166575, 0…
## $ postnap_change_uncued <dbl> 0.25266550, -0.39671291, 0.30253356, 1.89179922…
## $ week_change_cued <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ week_change_uncued <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ diff_biaschange_cued <dbl> 0.371668148, -0.359624732, -0.192821041, -0.570…
## $ diff_biaschange_uncued <dbl> -0.07323769, 0.65466998, 0.81248335, -0.0710435…
## $ diff_biaschange <dbl> 0.44490584, -1.01429471, -1.00530440, -0.499230…
## $ base_IAT_race <dbl> 0.57544182, 0.09911241, 1.52435622, 0.13108478,…
## $ base_IAT_gen <dbl> 0.60953653, 0.64396538, 0.20577365, 0.35314196,…
## $ pre_IAT_race <dbl> 0.55905291, -0.13380639, 0.37990232, -0.9420955…
## $ pre_IAT_gen <dbl> 0.21462144, 0.33985028, 0.51077026, -0.02933191…
## $ post_IAT_race <dbl> 0.68191015, 0.04463480, 0.68243589, 0.94970369,…
## $ week_IAT_race <dbl> 0.20377367, 0.45873715, 0.71187286, 0.20212832,…
## $ post_IAT_gen <dbl> 0.467286940, -0.056862624, -0.002583615, -0.245…
## $ week_IAT_gen <dbl> 0.68277422, -0.01070460, 0.39859469, 0.92341592…
## $ filter_. <chr> "Selected", "Selected", "Selected", "Selected",…
## $ cues_total <dbl> 142.5, 180.0, 232.5, 240.0, 225.0, 375.0, 345.0…
## $ baselineIAT <dbl> 0.5924892, 0.3715389, 0.8650649, 0.2421134, 0.3…
## $ prenapIAT <dbl> 0.38683717, 0.10302195, 0.44533629, -0.48571372…
## $ postnapIAT <dbl> 0.57459854, -0.00611391, 0.33992614, 0.35185714…
## $ weekIAT <dbl> 0.44327395, 0.22401627, 0.55523378, 0.56277212,…Now, I have to repeat the process to find means and standard errors:
0 years at Uni bias change:
Select participants only attend uni 0 years:
year0uni_participants <- mutated_exploratorydata1 %>%
filter(General_1_UniYears == "0")calculate avg bias levels:
year0uni <- year0uni_participants %>%
select(baselineIAT, prenapIAT, postnapIAT, weekIAT)
year0uni_avgbias <- year0uni %>%
summarise(year0uni, mean = "mean")
head(year0uni_avgbias)## # A tibble: 6 x 5
## baselineIAT prenapIAT postnapIAT weekIAT mean
## <dbl> <dbl> <dbl> <dbl> <chr>
## 1 0.310 0.0320 0.0751 0.0560 mean
## 2 0.867 0.122 0.806 -0.173 mean
## 3 0.646 0.611 0.535 0.142 mean
## 4 0.805 0.305 -0.627 0.230 mean
## 5 0.418 0.624 0.0990 0.0556 mean
## 6 0.564 0.403 0.262 0.750 meanyear0uni_se <- std.error(year0uni)
head(year0uni_se)## baselineIAT prenapIAT postnapIAT weekIAT
## 0.08727692 0.06171902 0.11709737 0.098728381 year at Uni bias change:
Select participants only attend uni 1 years:
year1uni_participants <- mutated_exploratorydata1 %>%
filter(General_1_UniYears == "1")
print(year1uni_participants)## # A tibble: 4 x 105
## # Rowwise:
## ParticipantID exclude cue_presented heard_cue_report heard_cue_exit
## <chr> <chr> <chr> <chr> <chr>
## 1 ub25 no yes no unsure
## 2 ub31 no yes no no
## 3 ub34 no yes no no
## 4 ub45 no yes no no
## # … with 100 more variables: predicted_cue <chr>, Cue_condition <chr>,
## # Counterbias_order <chr>, Sound_assignment <chr>, IAT1_order <chr>,
## # IAT234_order <chr>, IAT_order <chr>, compensation <chr>,
## # General_1_Age <int>, General_1_Sex <chr>, General_1_Race <chr>,
## # General_1_English <chr>, General_1_EnglishYrs <int>,
## # General_1_Caffeine <chr>, General_1_CaffCups <int>,
## # General_1_CaffHrsAgo <dbl>, General_1_SleepDisor <chr>,
## # General_1_MentalDiso <chr>, General_1_Meds <chr>, General_1_MedList <chr>,
## # General_1_University <chr>, General_1_UniYears <int>, Demo_1_Ethnic <chr>,
## # Demo_1_Racial <chr>, Demo_1_Gender <chr>, Demo_1_NonParticipat <chr>,
## # Epworth_1_Read <chr>, Epworth_1_TV <chr>, Epworth_1_Public <chr>,
## # Epworth_1_Passenger <chr>, Epworth_1_LyingDown <chr>,
## # Epworth_1_Talking <chr>, Epworth_1_Lunch <chr>, Epworth_1_Traffic <chr>,
## # Epworth_total <int>, AlertTest_1_Concentr_1 <int>,
## # AlertTest_1_Refresh_1 <int>, AlertTest_1_Feel <chr>,
## # AlertTest_2_Concentr_1 <int>, AlertTest_2_Refresh_1 <int>,
## # AlertTest_2_Feel <chr>, AlertTest_3_Concentr_1 <int>,
## # AlertTest_3_Refresh_1 <int>, AlertTest_3_Feel <chr>,
## # AlertTest_4_Concentr_1 <int>, AlertTest_4_Refresh_1 <int>,
## # AlertTest_4_Feel <chr>, S1_ExitQ_1_sound <chr>,
## # S1_ExitQ_1_soundaffect <chr>, S1_ExitQ_2_sound <chr>,
## # S1_ExitQ_3_sound <chr>, S1_ExitQ_4_sound <chr>,
## # S1_ExitQ_4_soundaffect <chr>, S1_ExitQ_5_sound <chr>,
## # S1_ExitQ_5_soundaffect <chr>, S2_ExitQ_1_sound <chr>,
## # S2_ExitQ_1_soundaffect <chr>, S2_ExitQ_2_sound <chr>,
## # S2_ExitQ_3_sound <chr>, S2_ExitQ_4_sound <chr>,
## # S2_ExitQ_4_soundaffect <chr>, S2_ExitQ_5_sound <chr>,
## # S2_ExitQ_5_soundaffect <chr>, Total_sleep <int>, Wake_amount <int>,
## # NREM1_amount <int>, NREM2_amount <dbl>, SWS_amount <int>, REM_amount <int>,
## # SWSxREM <int>, cue_minutes <dbl>, baseIATcued <dbl>, baseIATuncued <dbl>,
## # preIATcued <dbl>, preIATuncued <dbl>, postIATcued <dbl>,
## # postIATuncued <dbl>, weekIATcued <dbl>, weekIATuncued <dbl>,
## # postnap_change_cued <dbl>, postnap_change_uncued <dbl>,
## # week_change_cued <int>, week_change_uncued <int>,
## # diff_biaschange_cued <dbl>, diff_biaschange_uncued <dbl>,
## # diff_biaschange <dbl>, base_IAT_race <dbl>, base_IAT_gen <dbl>,
## # pre_IAT_race <dbl>, pre_IAT_gen <dbl>, post_IAT_race <dbl>,
## # week_IAT_race <dbl>, post_IAT_gen <dbl>, week_IAT_gen <dbl>,
## # filter_. <chr>, cues_total <dbl>, baselineIAT <dbl>, prenapIAT <dbl>,
## # postnapIAT <dbl>, weekIAT <dbl>calculate avg bias levels:
year1uni <- year1uni_participants %>%
select(baselineIAT, prenapIAT, postnapIAT, weekIAT)
year1uni_avgbias <- year1uni %>%
summarise(across(contains("IAT"), list(mean = mean)))
head(year1uni_avgbias)## # A tibble: 4 x 4
## baselineIAT_mean prenapIAT_mean postnapIAT_mean weekIAT_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.665 0.906 0.375 0.433
## 2 0.527 0.216 0.891 0.765
## 3 0.401 -0.0223 0.0634 0.291
## 4 0.909 0.291 0.160 0.516year1uni_se <- std.error(year1uni)
head(year1uni_se)## baselineIAT prenapIAT postnapIAT weekIAT
## 0.1087993 0.1977530 0.1848361 0.09931042 years at Uni bias change:
Select participants only attend uni 2 years:
year2uni_participants <- mutated_exploratorydata1 %>%
filter(General_1_UniYears == "2")
print(year2uni_participants)## # A tibble: 10 x 105
## # Rowwise:
## ParticipantID exclude cue_presented heard_cue_report heard_cue_exit
## <chr> <chr> <chr> <chr> <chr>
## 1 ub8 no yes no no
## 2 ub13 no yes no no
## 3 ub32 no yes maybe, unsure, unclear no
## 4 ub36 no yes no no
## 5 ub38 no yes no no
## 6 ub40 no yes no no
## 7 ub41 no yes no no
## 8 ub43 no yes no no
## 9 ub44 no yes no no
## 10 ub48 no yes no no
## # … with 100 more variables: predicted_cue <chr>, Cue_condition <chr>,
## # Counterbias_order <chr>, Sound_assignment <chr>, IAT1_order <chr>,
## # IAT234_order <chr>, IAT_order <chr>, compensation <chr>,
## # General_1_Age <int>, General_1_Sex <chr>, General_1_Race <chr>,
## # General_1_English <chr>, General_1_EnglishYrs <int>,
## # General_1_Caffeine <chr>, General_1_CaffCups <int>,
## # General_1_CaffHrsAgo <dbl>, General_1_SleepDisor <chr>,
## # General_1_MentalDiso <chr>, General_1_Meds <chr>, General_1_MedList <chr>,
## # General_1_University <chr>, General_1_UniYears <int>, Demo_1_Ethnic <chr>,
## # Demo_1_Racial <chr>, Demo_1_Gender <chr>, Demo_1_NonParticipat <chr>,
## # Epworth_1_Read <chr>, Epworth_1_TV <chr>, Epworth_1_Public <chr>,
## # Epworth_1_Passenger <chr>, Epworth_1_LyingDown <chr>,
## # Epworth_1_Talking <chr>, Epworth_1_Lunch <chr>, Epworth_1_Traffic <chr>,
## # Epworth_total <int>, AlertTest_1_Concentr_1 <int>,
## # AlertTest_1_Refresh_1 <int>, AlertTest_1_Feel <chr>,
## # AlertTest_2_Concentr_1 <int>, AlertTest_2_Refresh_1 <int>,
## # AlertTest_2_Feel <chr>, AlertTest_3_Concentr_1 <int>,
## # AlertTest_3_Refresh_1 <int>, AlertTest_3_Feel <chr>,
## # AlertTest_4_Concentr_1 <int>, AlertTest_4_Refresh_1 <int>,
## # AlertTest_4_Feel <chr>, S1_ExitQ_1_sound <chr>,
## # S1_ExitQ_1_soundaffect <chr>, S1_ExitQ_2_sound <chr>,
## # S1_ExitQ_3_sound <chr>, S1_ExitQ_4_sound <chr>,
## # S1_ExitQ_4_soundaffect <chr>, S1_ExitQ_5_sound <chr>,
## # S1_ExitQ_5_soundaffect <chr>, S2_ExitQ_1_sound <chr>,
## # S2_ExitQ_1_soundaffect <chr>, S2_ExitQ_2_sound <chr>,
## # S2_ExitQ_3_sound <chr>, S2_ExitQ_4_sound <chr>,
## # S2_ExitQ_4_soundaffect <chr>, S2_ExitQ_5_sound <chr>,
## # S2_ExitQ_5_soundaffect <chr>, Total_sleep <int>, Wake_amount <int>,
## # NREM1_amount <int>, NREM2_amount <dbl>, SWS_amount <int>, REM_amount <int>,
## # SWSxREM <int>, cue_minutes <dbl>, baseIATcued <dbl>, baseIATuncued <dbl>,
## # preIATcued <dbl>, preIATuncued <dbl>, postIATcued <dbl>,
## # postIATuncued <dbl>, weekIATcued <dbl>, weekIATuncued <dbl>,
## # postnap_change_cued <dbl>, postnap_change_uncued <dbl>,
## # week_change_cued <int>, week_change_uncued <int>,
## # diff_biaschange_cued <dbl>, diff_biaschange_uncued <dbl>,
## # diff_biaschange <dbl>, base_IAT_race <dbl>, base_IAT_gen <dbl>,
## # pre_IAT_race <dbl>, pre_IAT_gen <dbl>, post_IAT_race <dbl>,
## # week_IAT_race <dbl>, post_IAT_gen <dbl>, week_IAT_gen <dbl>,
## # filter_. <chr>, cues_total <dbl>, baselineIAT <dbl>, prenapIAT <dbl>,
## # postnapIAT <dbl>, weekIAT <dbl>calculate avg bias levels:
year2uni <- year2uni_participants %>%
select(baselineIAT, prenapIAT, postnapIAT, weekIAT)
year2uni_avgbias <- year2uni %>%
summarise(across(contains("IAT"), list(mean = mean)))
head(year2uni_avgbias)## # A tibble: 6 x 4
## baselineIAT_mean prenapIAT_mean postnapIAT_mean weekIAT_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.865 0.445 0.340 0.555
## 2 0.606 0.197 0.624 0.839
## 3 0.442 -0.378 0.311 0.950
## 4 0.503 0.398 0.199 0.631
## 5 -0.219 -0.273 0.0524 0.0274
## 6 0.437 -0.176 0.269 0.402year2uni_se <- std.error(year2uni)
head(year2uni_se)## baselineIAT prenapIAT postnapIAT weekIAT
## 0.1082313 0.1509254 0.1050117 0.11071713 years at Uni bias change:
Select participants only attend uni 3 years:
year3uni_participants <- mutated_exploratorydata1 %>%
filter(General_1_UniYears == "3")
print(year3uni_participants)## # A tibble: 6 x 105
## # Rowwise:
## ParticipantID exclude cue_presented heard_cue_report heard_cue_exit
## <chr> <chr> <chr> <chr> <chr>
## 1 ub6 no yes no no
## 2 ub7 no yes no unsure
## 3 ub9 no yes no no
## 4 ub35 no yes no no
## 5 ub42 no yes no no
## 6 ub49 no yes no no
## # … with 100 more variables: predicted_cue <chr>, Cue_condition <chr>,
## # Counterbias_order <chr>, Sound_assignment <chr>, IAT1_order <chr>,
## # IAT234_order <chr>, IAT_order <chr>, compensation <chr>,
## # General_1_Age <int>, General_1_Sex <chr>, General_1_Race <chr>,
## # General_1_English <chr>, General_1_EnglishYrs <int>,
## # General_1_Caffeine <chr>, General_1_CaffCups <int>,
## # General_1_CaffHrsAgo <dbl>, General_1_SleepDisor <chr>,
## # General_1_MentalDiso <chr>, General_1_Meds <chr>, General_1_MedList <chr>,
## # General_1_University <chr>, General_1_UniYears <int>, Demo_1_Ethnic <chr>,
## # Demo_1_Racial <chr>, Demo_1_Gender <chr>, Demo_1_NonParticipat <chr>,
## # Epworth_1_Read <chr>, Epworth_1_TV <chr>, Epworth_1_Public <chr>,
## # Epworth_1_Passenger <chr>, Epworth_1_LyingDown <chr>,
## # Epworth_1_Talking <chr>, Epworth_1_Lunch <chr>, Epworth_1_Traffic <chr>,
## # Epworth_total <int>, AlertTest_1_Concentr_1 <int>,
## # AlertTest_1_Refresh_1 <int>, AlertTest_1_Feel <chr>,
## # AlertTest_2_Concentr_1 <int>, AlertTest_2_Refresh_1 <int>,
## # AlertTest_2_Feel <chr>, AlertTest_3_Concentr_1 <int>,
## # AlertTest_3_Refresh_1 <int>, AlertTest_3_Feel <chr>,
## # AlertTest_4_Concentr_1 <int>, AlertTest_4_Refresh_1 <int>,
## # AlertTest_4_Feel <chr>, S1_ExitQ_1_sound <chr>,
## # S1_ExitQ_1_soundaffect <chr>, S1_ExitQ_2_sound <chr>,
## # S1_ExitQ_3_sound <chr>, S1_ExitQ_4_sound <chr>,
## # S1_ExitQ_4_soundaffect <chr>, S1_ExitQ_5_sound <chr>,
## # S1_ExitQ_5_soundaffect <chr>, S2_ExitQ_1_sound <chr>,
## # S2_ExitQ_1_soundaffect <chr>, S2_ExitQ_2_sound <chr>,
## # S2_ExitQ_3_sound <chr>, S2_ExitQ_4_sound <chr>,
## # S2_ExitQ_4_soundaffect <chr>, S2_ExitQ_5_sound <chr>,
## # S2_ExitQ_5_soundaffect <chr>, Total_sleep <int>, Wake_amount <int>,
## # NREM1_amount <int>, NREM2_amount <dbl>, SWS_amount <int>, REM_amount <int>,
## # SWSxREM <int>, cue_minutes <dbl>, baseIATcued <dbl>, baseIATuncued <dbl>,
## # preIATcued <dbl>, preIATuncued <dbl>, postIATcued <dbl>,
## # postIATuncued <dbl>, weekIATcued <dbl>, weekIATuncued <dbl>,
## # postnap_change_cued <dbl>, postnap_change_uncued <dbl>,
## # week_change_cued <int>, week_change_uncued <int>,
## # diff_biaschange_cued <dbl>, diff_biaschange_uncued <dbl>,
## # diff_biaschange <dbl>, base_IAT_race <dbl>, base_IAT_gen <dbl>,
## # pre_IAT_race <dbl>, pre_IAT_gen <dbl>, post_IAT_race <dbl>,
## # week_IAT_race <dbl>, post_IAT_gen <dbl>, week_IAT_gen <dbl>,
## # filter_. <chr>, cues_total <dbl>, baselineIAT <dbl>, prenapIAT <dbl>,
## # postnapIAT <dbl>, weekIAT <dbl>calculate avg bias levels:
year3uni <- year3uni_participants %>%
select(baselineIAT, prenapIAT, postnapIAT, weekIAT)
year3uni_avgbias <- year3uni %>%
summarise(across(contains("IAT"), list(mean = mean)))
head(year3uni_avgbias)## # A tibble: 6 x 4
## baselineIAT_mean prenapIAT_mean postnapIAT_mean weekIAT_mean
## <dbl> <dbl> <dbl> <dbl>
## 1 0.592 0.387 0.575 0.443
## 2 0.372 0.103 -0.00611 0.224
## 3 0.242 -0.486 0.352 0.563
## 4 0.924 0.453 0.204 0.538
## 5 0.740 0.508 0.0421 -0.467
## 6 0.387 0.0875 0.376 0.249year3uni_se <- std.error(year3uni)
head(year3uni_se)## baselineIAT prenapIAT postnapIAT weekIAT
## 0.1049749 0.1508470 0.0898536 0.1562415Create dataframe
When creating this dataframe, it kept coming up with an error, saying that there are differing number of rows. By this stage, I decided to just wait for the Q&A session on Tuesday and see if there was anything that could be used during that session
biasdata <- data.frame(
condition = factor(c("0", "0", "0", "0", "1", "1", "1", "1", "2", "2", "2", "2", "3", "3", "3", "3")),
time = factor(c("Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week")),
levels = c("Baseline", "Prenap", "Postnap", "1-week"),
bias_av = c(year0uni_avgbias$baselineIAT_mean, year0uni_avgbias$prenapIAT_mean, year0uni_avgbias$postnapIAT_mean, year0uni_avgbias$weekIAT_mean,
year1uni_avgbias$baselineIAT_mean, year1uni_avgbias$prenapIAT_mean, year1uni_avgbias$postnapIAT_mean, year1uni_avgbias$weekIAT_mean,
year2uni_avgbias$baselineIAT_mean, year2uni_avgbias$prenapIAT_mean, year2uni_avgbias$postnapIAT_mean, year2uni_avgbias$weekIAT_mean,
year3uni_avgbias$baselineIAT_mean, year3uni_avgbias$prenapIAT_mean, year3uni_avgbias$postnapIAT_mean, year3uni_avgbias$weekIAT_mean))## Warning: Unknown or uninitialised column: `baselineIAT_mean`.## Warning: Unknown or uninitialised column: `prenapIAT_mean`.## Warning: Unknown or uninitialised column: `postnapIAT_mean`.## Warning: Unknown or uninitialised column: `weekIAT_mean`.Attempt 4
After the Q&A where Jenny went through how to do descriptive statistics quickly and easily, I tried to do the same:
I decided to create the means and standard errors for each of the 4 IAT timepoints, grouped by the number of years at university. I also tried calculating standard error, using the formula rather than an actual function:
year_summary_baseline <- mutated_exploratorydata1 %>%
group_by(General_1_UniYears) %>%
summarise(mean = mean(baselineIAT),
sd = sd(baselineIAT),
n = n(),
se = sd/sqrt(n))
year_summary_prenap <- mutated_exploratorydata1 %>%
group_by(General_1_UniYears) %>%
summarise(mean = mean(prenapIAT),
sd = sd(prenapIAT),
n = n(),
se = sd/sqrt(n))
year_summary_postnap <- mutated_exploratorydata1 %>%
group_by(General_1_UniYears) %>%
summarise(mean = mean(postnapIAT),
sd = sd(postnapIAT),
n = n(),
se = sd/sqrt(n))
year_summary_week <- mutated_exploratorydata1 %>%
group_by(General_1_UniYears) %>%
summarise(mean = mean(weekIAT),
sd = sd(weekIAT),
n = n(),
se = sd/sqrt(n))Now, that the means and standard errors have been calculated, I have to put the descriptive statistics into a table for each timepoint (i.e. create 4 tables). I used the gt() package to create the table.
year_summary_baseline %>%
gt() %>%
tab_header(title = md("**Baseline IAT timepoint**")) %>%
fmt_number(
columns = vars(mean, sd, se),
decimals = 2
) %>%
cols_label(General_1_UniYears = "Number of Years at University",
mean = "Mean",
sd = "SD",
n = "n",
se = "SE")## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead
## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead| Baseline IAT timepoint | ||||
|---|---|---|---|---|
| Number of Years at University | Mean | SD | n | SE |
| 0 | 0.56 | 0.29 | 11 | 0.09 |
| 1 | 0.63 | 0.22 | 4 | 0.11 |
| 2 | 0.53 | 0.34 | 10 | 0.11 |
| 3 | 0.54 | 0.26 | 6 | 0.10 |
year_summary_prenap %>%
gt() %>%
tab_header(title = md("**Prenap IAT timepoint**")) %>%
fmt_number(
columns = vars(mean, sd, se),
decimals = 2
) %>%
cols_label(General_1_UniYears = "Number of Years at University",
mean = "Mean",
sd = "SD",
n = "n",
se = "SE")## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead
## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead| Prenap IAT timepoint | ||||
|---|---|---|---|---|
| Number of Years at University | Mean | SD | n | SE |
| 0 | 0.34 | 0.20 | 11 | 0.06 |
| 1 | 0.35 | 0.40 | 4 | 0.20 |
| 2 | 0.18 | 0.48 | 10 | 0.15 |
| 3 | 0.18 | 0.37 | 6 | 0.15 |
year_summary_postnap %>%
gt() %>%
tab_header(title = md("**Postnap IAT timepoint**")) %>%
fmt_number(
columns = vars(mean, sd, se),
decimals = 2
) %>%
cols_label(General_1_UniYears = "Number of Years at University",
mean = "Mean",
sd = "SD",
n = "n",
se = "SE")## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead
## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead| Postnap IAT timepoint | ||||
|---|---|---|---|---|
| Number of Years at University | Mean | SD | n | SE |
| 0 | 0.25 | 0.39 | 11 | 0.12 |
| 1 | 0.37 | 0.37 | 4 | 0.18 |
| 2 | 0.29 | 0.33 | 10 | 0.11 |
| 3 | 0.26 | 0.22 | 6 | 0.09 |
year_summary_week %>%
gt() %>%
tab_header(title = md("**One-week delay IAT timepoint**")) %>%
fmt_number(
columns = vars(mean, sd, se),
decimals = 2
) %>%
cols_label(General_1_UniYears = "Number of Years at University",
mean = "Mean",
sd = "SD",
n = "n",
se = "SE")## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead
## Warning: `columns = vars(...)` has been deprecated in gt 0.3.0:
## * please use `columns = c(...)` instead| One-week delay IAT timepoint | ||||
|---|---|---|---|---|
| Number of Years at University | Mean | SD | n | SE |
| 0 | 0.30 | 0.33 | 11 | 0.10 |
| 1 | 0.50 | 0.20 | 4 | 0.10 |
| 2 | 0.56 | 0.35 | 10 | 0.11 |
| 3 | 0.26 | 0.38 | 6 | 0.16 |
Visualisation
Now that the descriptive statistics have been calculated, it's time to create the figure. First, I have to create the dataframe for the figure. - condition = defines the data points for each of the 4 lines in the graph; one for each year spent in uni - time = defines the x-axis, and the 4 different IAT timepoints - level also helps define the x-axis - bias_av defines the data points for the average bias scores - stderror defines the data points for the standard errors
biasdata <- data.frame(
condition = factor(c("0", "0", "0", "0", "1", "1", "1", "1", "2", "2", "2", "2", "3", "3", "3", "3")),
time = factor(c("Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week", "Baseline", "Prenap", "Postnap", "1-week")),
levels = c("Baseline", "Prenap", "Postnap", "1-week"),
bias_av = c(0.5633872, 0.3387359, 0.2454916, 0.2977948,
0.6255864, 0.3477496, 0.3722320, 0.5012622,
0.5295660, 0.1786688, 0.2875954, 0.5551314,
0.5428986, 0.1754848, 0.2571501, 0.2583117),
stderror = c(0.08727692, 0.06171902, 0.1170974, 0.09872838,
0.6255864, 0.19775303, 0.1848361, 0.09931040,
0.10823134, 0.15092541, 0.1050117, 0.11071705,
0.10497492, 0.15084704, 0.0898536, 0.15624148
))
head(biasdata)## condition time levels bias_av stderror
## 1 0 Baseline Baseline 0.5633872 0.08727692
## 2 0 Prenap Prenap 0.3387359 0.06171902
## 3 0 Postnap Postnap 0.2454916 0.11709740
## 4 0 1-week 1-week 0.2977948 0.09872838
## 5 1 Baseline Baseline 0.6255864 0.62558640
## 6 1 Prenap Prenap 0.3477496 0.19775303Attempt 1
Now that the dataframe is created, the figure must be formatted. I used ggplot() to create the figure. However, it only came up with standard error bars for some reason. The lines to connect each of the time conditions aren't appearing and the legend for the 4 different conditions is missing.
fig_1 <- ggplot(data = biasdata,
aes(x = time, y = bias_av, fill = condition))+
geom_line()+
geom_errorbar(aes(x = time, ymin = bias_av - stderror, ymax = bias_av + stderror),
width=0.1,
colour="grey",
alpha= 0.9) +
ylim(0.0, 0.7) +
labs(x = "",
y = "D600 Bias Score",
caption = "Fig 3. Average D600 scores at each IAT timepoint") +
theme_bw()
print(fig_1)## geom_path: Each group consists of only one observation. Do you need to adjust
## the group aesthetic?
Attempt 2
- I tried defining the x-axis better, by first defining the variable to be used, and the levels too.
- The y-axis definitely needed to be defined better, first by
colour(changes the colour of each line by each condition) and then bygroup(allows for partitioning of the data by condition) - I also changed the theme to classic, because I didn't like the grid lines in the plot
fig_1.2 <- ggplot(data = biasdata,
aes(x = factor(time, level = c("Baseline", "Prenap", "Postnap", "1-week")),
y = bias_av,
colour = condition,
group = condition)) +
geom_line()+
geom_errorbar(aes(x = time, ymin = bias_av - stderror, ymax = bias_av + stderror),
width=0.1,
colour="grey",
alpha= 0.9) +
ylim(0.0, 0.7) +
labs(x = "",
y = "D600 Bias Score",
caption = "Fig 3. Average D600 scores at each IAT timepoint",
title = "Do the number of years spent in university affect students' bias scores?") +
theme_classic()
print(fig_1.2)
Attempt 3
However, I realised that the legend title for my plot didn't make sense. I first tried defining the title by using labs() and defining for group =, which didn't change anything. By defining for colour =, I was able to change the legend title.
fig_1.3 <- ggplot(data = biasdata,
aes(x = factor(time, level = c("Baseline", "Prenap", "Postnap", "1-week")),
y = bias_av,
colour = condition,
group = condition)) +
geom_line()+
geom_errorbar(aes(x = time, ymin = bias_av - stderror, ymax = bias_av + stderror),
width=0.1,
colour="grey",
alpha= 0.9) +
ylim(0.0, 0.7) +
labs(x = "",
y = "D600 Bias Score",
caption = "Fig 3. Average D600 scores at each IAT timepoint",
title = "Do the number of years spent in university affect students' bias scores?",
colour = "Number of years in university") +
theme_classic()
print(fig_1.3)
Interestingly, it seems that those who spent 1 year in uni are the least affected by the TMR procedure, maintaining a relatively high bias score throughout the 4 timepoints. Meanwhile, 2- or 3- year students have similar, low bias score before the nap but 2-year students have the fastest recovery, surpassing 1-year students at the one-week timepoint. In contrast, 0- and 3- years maintain low bias scores even one-week later, suggesting that the TMR procedure may have had a greater impact on them compared to the other year groups.
Statistics !
I have to compare means between conditions (number of years spent in university) at each IAT timepoint, so I thought I would use ANOVA. I used the method that Jenny showed in the Q&A session. However, I have no idea how to interpret it.
Timepoint 1: baseline
m1 <- aov(mean ~ General_1_UniYears, data = year_summary_baseline)
summary(m1)## Df Sum Sq Mean Sq F value Pr(>F)
## General_1_UniYears 1 0.001240 0.001240 0.594 0.522
## Residuals 2 0.004177 0.002088Timepoint 2: prenap
m2 <- aov(mean ~ General_1_UniYears, data = year_summary_prenap)
summary(m2)## Df Sum Sq Mean Sq F value Pr(>F)
## General_1_UniYears 1 0.021703 0.021703 7.291 0.114
## Residuals 2 0.005954 0.002977Timepoint 3: postnap
m3 <- aov(mean ~ General_1_UniYears, data = year_summary_postnap)
summary(m3)## Df Sum Sq Mean Sq F value Pr(>F)
## General_1_UniYears 1 0.000123 0.000123 0.025 0.888
## Residuals 2 0.009703 0.004852Timepoint 4: one-week delay
m4 <- aov(mean ~ General_1_UniYears, data = year_summary_week)
summary(m4)## Df Sum Sq Mean Sq F value Pr(>F)
## General_1_UniYears 1 0.00021 0.00021 0.006 0.943
## Residuals 2 0.06459 0.03230I'm not great with statistics, but I'm assuming Pr(>F) is the p-value associated with each F-statistic. Since all four p-values are greater than 0.05, there is no evidence to suggest that the number of years spent in university affect students' bias scores.
Challenges and successes
Most of these have been documented above, however I'm not sure if I did my statistics right. Hopefully, the next Q&A session, I'll have a chance to ask questions regarding this.
Next steps
I now need to decide on which two other questions I'll be including in my verification report, and then carrying out the coding to answer them. I also need to research more about the statistics part of this, as statistics is definitely not one of my strengths.