DATA 606 Data Project Proposal
Amber Ferger
Data Preparation
First, I’ll load in the data and rename the columns.
library(dplyr)
library(tidyverse)
religionData <- read.csv('religionData.csv', header= TRUE, sep = ',')
religionData %>% as_tibble(religionData)
# rename columns
colNames <- c('RELIGION','RELIGION2', 'EVANGELICAL', 'RELIGIOUS_SERVICES', 'FREQ_PRAY_WITH_MOTIONS', 'FREQ_PRAY_WITH_OBJECTS', 'FREQ_PRAY_BEFORE_MEALS',
'FREQ_PRAY_FOR_OTHERS', 'FREQ_ASK_TO_PRAY_WITH_SOMEONE', 'FREQ_BRING_UP_RELIGION',
'FREQ_ASK_ABOUT_RELIGION', 'FREQ_DECLINE_FOOD_FOR_RELIGION', 'FREQ_WEAR_RELIGIOUS_CLOTHING', 'FREQ_PARTICIPATE_IN_PUBLIC_RELIGIOUS_EVENT',
'COMFORT_OWN_PRAY_WITH_MOTIONS',
'COMFORT_OWN_PRAY_WITH_OBJECTS',
'COMFORT_OWN_PRAY_BEFORE_MEALS',
'COMFORT_OWN_PRAY_FOR_OTHERS',
'COMFORT_OWN_ASK_TO_PRAY_WITH_SOMEONE',
'COMFORT_OWN_BRING_UP_RELIGION',
'COMFORT_OWN_ASK_ABOUT_RELIGION',
'COMFORT_OWN_DECLINE_FOOD_FOR_RELIGION',
'COMFORT_OWN_WEAR_RELIGIOUS_CLOTHING',
'COMFORT_OWN_PARTICIPATE_IN_PUBLIC_RELIGIOUS_EVENT',
'COMFORT_OTHER_PRAY_WITH_MOTIONS', 'COMFORT_OTHER_PRAY_WITH_OBJECTS', 'COMFORT_OTHER_PRAY_BEFORE_MEALS', 'COMFORT_OTHER_PRAY_FOR_OTHERS', 'COMFORT_OTHER_ASK_TO_PRAY_WITH_SOMEONE', 'COMFORT_OTHER_BRING_UP_RELIGION', 'COMFORT_OTHER_ASK_ABOUT_RELIGION', 'COMFORT_OTHER_DECLINE_FOOD_FOR_RELIGION',
'COMFORT_OTHER_WEAR_RELIGIOUS_CLOTHING', 'COMFORT_OTHER_PARTICIPATE_IN_PUBLIC_RELIGIOUS_EVENT', 'COMFORT_SEE_OTHER_PRAY_WITH_MOTIONS', 'COMFORT_SEE_OTHER_PRAY_WITH_OBJECTS', 'COMFORT_SEE_OTHER_PRAY_BEFORE_MEALS', 'COMFORT_SEE_OTHER_PRAY_FOR_OTHERS', 'COMFORT_SEE_OTHER_ASK_TO_PRAY_WITH_SOMEONE', 'COMFORT_SEE_OTHER_BRING_UP_RELIGION',
'COMFORT_SEE_OTHER_ASK_ABOUT_RELIGION', 'COMFORT_SEE_OTHER_DECLINE_FOOD_FOR_RELIGION', 'COMFORT_SEE_OTHER_WEAR_RELIGIOUS_CLOTHING', 'COMFORT_SEE_OTHER_PARTICIPATE_IN_PUBLIC_RELIGIOUS_EVENT', 'AGE', 'GENDER', 'HOUSEHOLD_SALARY', 'US_REGION')
names(religionData) <- colNamesThere are a number of fields within this dataset, so I am going to subset it only to general demographics and survey responses related to the comfort of seeing religious actions outside of the respondent’s religion.
colsToKeep <- c('COMFORT_SEE_OTHER_PRAY_WITH_MOTIONS', 'COMFORT_SEE_OTHER_PRAY_WITH_OBJECTS', 'COMFORT_SEE_OTHER_PRAY_BEFORE_MEALS', 'COMFORT_SEE_OTHER_PRAY_FOR_OTHERS', 'COMFORT_SEE_OTHER_ASK_TO_PRAY_WITH_SOMEONE', 'COMFORT_SEE_OTHER_BRING_UP_RELIGION',
'COMFORT_SEE_OTHER_ASK_ABOUT_RELIGION', 'COMFORT_SEE_OTHER_DECLINE_FOOD_FOR_RELIGION', 'COMFORT_SEE_OTHER_WEAR_RELIGIOUS_CLOTHING', 'COMFORT_SEE_OTHER_PARTICIPATE_IN_PUBLIC_RELIGIOUS_EVENT', 'AGE', 'GENDER', 'HOUSEHOLD_SALARY', 'US_REGION')
religionData <- religionData %>%
filter(RELIGION != 'Response') %>%
select(colsToKeep)In order to quantify each individual’s comfort with public religious displays, I need to convert my categorical features to numeric. First, I will rank each survey response in order of comfort (with Not at all comfortable having the lowest ranking and Extremely comfortable having the highest ranking)
comfortColumns <- seq(1:10)
for (i in comfortColumns) {
religionData[[i]] <- factor(religionData[[i]], levels = c("", "Response", "Not at all comfortable", "Not so comfortable", "Somewhat comfortably", "Very comfortable", "Extremely comfortable"), ordered = TRUE)
}
# eliminate records where any of the survey responses are blank
toBeRemoved<-which(religionData$COMFORT_SEE_OTHER_PRAY_WITH_MOTIONS==""|religionData$COMFORT_SEE_OTHER_PRAY_WITH_OBJECTS==""|religionData$COMFORT_SEE_OTHER_PRAY_BEFORE_MEALS==""|religionData$COMFORT_SEE_OTHER_PRAY_FOR_OTHERS==""|religionData$COMFORT_SEE_OTHER_ASK_TO_PRAY_WITH_SOMEONE==""|religionData$COMFORT_SEE_OTHER_BRING_UP_RELIGION==""|religionData$COMFORT_SEE_OTHER_ASK_ABOUT_RELIGION==""|religionData$COMFORT_SEE_OTHER_DECLINE_FOOD_FOR_RELIGION==""|religionData$COMFORT_SEE_OTHER_WEAR_RELIGIOUS_CLOTHING==""|religionData$COMFORT_SEE_OTHER_PARTICIPATE_IN_PUBLIC_RELIGIOUS_EVENT=="")
religionData<-religionData[-toBeRemoved,]The survey responses are still categorical, so I need to convert them to numeric values. Since the columns are already ordered, the numerical values should maintain the hierarchy. I’ll also need to compute an average ranking across survey responses. I’ll create a new variable called AVERAGE_RANKING with this information.
religionData$AVERAGE_RATING <- 0
# loop through all survey questions and convert each response to a number
# add number to the AVERAGE_RATING column
for (i in comfortColumns) {
religionData[[i]]<-as.numeric(religionData[[i]])
religionData$AVERAGE_RATING <- religionData$AVERAGE_RATING + religionData[[i]]
}
# final average rating
religionData$AVERAGE_RATING <- religionData$AVERAGE_RATING/10Research question
You should phrase your research question in a way that matches up with the scope of inference your dataset allows for.
Are age, income, and region predictive of an individual’s comfort level with public religious displays?
Cases
What are the cases, and how many are there?
Each case represents a respondent’s survey answers. There are a total of 979 cases in the dataset.
Data collection
Describe the method of data collection.
This data was collected using a SurveyMonkey poll, conducted between July 29 and August 1, 2016. The survey asked 661 respondents questions about public displays of religion.
Type of study
What type of study is this (observational/experiment)?
This is a survey, which is a type of observational study.
Data Source
If you collected the data, state self-collected. If not, provide a citation/link.
This data came from: https://github.com/fivethirtyeight/data/tree/master/religion-survey.
Dependent Variable
What is the response variable? Is it quantitative or qualitative?
The response variable is comfort level and it is numeric.
Independent Variable
You should have two independent variables, one quantitative and one qualitative.
- Quantitative Variable - Age, Income
- Qualitative Variable - Region
Relevant summary statistics
Provide summary statistics for each the variables. Also include appropriate visualizations related to your research question (e.g. scatter plot, boxplots, etc). This step requires the use of R, hence a code chunk is provided below. Insert more code chunks as needed.
summary(religionData$AVERAGE_RATING)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.000 5.000 5.700 5.597 6.200 7.000table(religionData$AGE)##
## 18 - 29 30 - 44 45 - 59 60+ Response
## 212 255 264 248 0table(religionData$HOUSEHOLD_SALARY)##
## $0 to $9,999 $10,000 to $24,999 $100,000 to $124,999
## 86 93 87
## $125,000 to $149,999 $150,000 to $174,999 $175,000 to $199,999
## 52 37 15
## $200,000 and up $25,000 to $49,999 $50,000 to $74,999
## 53 166 151
## $75,000 to $99,999 Prefer not to answer Response
## 111 128 0table(religionData$US_REGION)##
## East North Central East South Central
## 13 164 50
## Middle Atlantic Mountain New England
## 124 67 65
## Pacific Response South Atlantic
## 146 0 191
## West North Central West South Central
## 68 91religionData %>%
group_by(AGE) %>%
summarise(MEAN_BY_AGE = mean(AVERAGE_RATING),
MEDIAN_BY_AGE = median(AVERAGE_RATING),
STDEV_BY_AGE = sd(AVERAGE_RATING))## # A tibble: 4 x 4
## AGE MEAN_BY_AGE MEDIAN_BY_AGE STDEV_BY_AGE
## <fct> <dbl> <dbl> <dbl>
## 1 18 - 29 5.67 5.75 0.966
## 2 30 - 44 5.54 5.6 1.02
## 3 45 - 59 5.61 5.7 0.938
## 4 60+ 5.59 5.6 0.808boxplot(religionData$AVERAGE_RATING~religionData$AGE)
religionData %>%
group_by(HOUSEHOLD_SALARY) %>%
summarise(MEAN_BY_SALARY = mean(AVERAGE_RATING),
MEDIAN_BY_SALARY = median(AVERAGE_RATING),
STDEV_BY_SALARY = sd(AVERAGE_RATING))## # A tibble: 11 x 4
## HOUSEHOLD_SALARY MEAN_BY_SALARY MEDIAN_BY_SALARY STDEV_BY_SALARY
## <fct> <dbl> <dbl> <dbl>
## 1 $0 to $9,999 5.54 5.75 1.16
## 2 $10,000 to $24,999 5.68 5.8 0.937
## 3 $100,000 to $124,999 5.54 5.6 0.939
## 4 $125,000 to $149,999 5.58 5.7 0.983
## 5 $150,000 to $174,999 5.45 5.6 1.01
## 6 $175,000 to $199,999 5.61 5.7 0.728
## 7 $200,000 and up 5.60 5.7 0.855
## 8 $25,000 to $49,999 5.63 5.8 1.01
## 9 $50,000 to $74,999 5.76 5.9 0.832
## 10 $75,000 to $99,999 5.47 5.6 0.945
## 11 Prefer not to answer 5.53 5.55 0.778boxplot(religionData$AVERAGE_RATING~religionData$HOUSEHOLD_SALARY)
religionData %>%
group_by(US_REGION) %>%
summarise(MEAN_BY_REGION = mean(AVERAGE_RATING),
MEDIAN_BY_REGION = median(AVERAGE_RATING),
STDEV_BY_REGION = sd(AVERAGE_RATING))## # A tibble: 10 x 4
## US_REGION MEAN_BY_REGION MEDIAN_BY_REGION STDEV_BY_REGION
## <fct> <dbl> <dbl> <dbl>
## 1 "" 5.19 5.3 1.07
## 2 East North Central 5.69 5.9 0.986
## 3 East South Central 5.85 5.95 0.735
## 4 Middle Atlantic 5.56 5.6 0.916
## 5 Mountain 5.41 5.5 1.05
## 6 New England 5.69 5.7 0.782
## 7 Pacific 5.59 5.7 1.02
## 8 South Atlantic 5.58 5.6 0.953
## 9 West North Central 5.51 5.65 0.880
## 10 West South Central 5.59 5.6 0.807boxplot(religionData$AVERAGE_RATING~religionData$US_REGION)
religionData %>%
group_by(GENDER) %>%
summarise(MEAN_BY_GENDER = mean(AVERAGE_RATING),
MEDIAN_BY_GENDER = median(AVERAGE_RATING),
STDEV_BY_GENDER = sd(AVERAGE_RATING))## # A tibble: 2 x 4
## GENDER MEAN_BY_GENDER MEDIAN_BY_GENDER STDEV_BY_GENDER
## <fct> <dbl> <dbl> <dbl>
## 1 Female 5.70 5.8 0.852
## 2 Male 5.49 5.6 1.01boxplot(religionData$AVERAGE_RATING~religionData$GENDER)
religionData %>%
group_by(AGE,GENDER) %>%
summarise(MEAN_BY_AGE_GENDER = mean(AVERAGE_RATING),
MEDIAN_BY_AGE_GENDER = median(AVERAGE_RATING),
STDEV_BY_AGE_GENDER = sd(AVERAGE_RATING))## # A tibble: 8 x 5
## # Groups: AGE [4]
## AGE GENDER MEAN_BY_AGE_GENDER MEDIAN_BY_AGE_GENDER STDEV_BY_AGE_GEND~
## <fct> <fct> <dbl> <dbl> <dbl>
## 1 18 - 29 Female 5.77 5.9 0.909
## 2 18 - 29 Male 5.55 5.5 1.01
## 3 30 - 44 Female 5.61 5.7 0.872
## 4 30 - 44 Male 5.46 5.5 1.16
## 5 45 - 59 Female 5.73 5.8 0.844
## 6 45 - 59 Male 5.48 5.65 1.02
## 7 60+ Female 5.68 5.7 0.794
## 8 60+ Male 5.47 5.55 0.813ggplot(religionData, aes(x=AVERAGE_RATING))+ geom_histogram()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.