[return to overview page]
Before delving deeper into the data, I will first give an overview of the data, so that you might get a better “feel” for the basic structure and nature of the responses (statements) that people provided. I will do this by going through each of the six questions to which people responded, and showing a sample set of responses for each question.
Before loading in any of the data, I will first load some very useful packages in R that will faciliated analysis and cleaning of the data.
# this is actually a series of packages, highly useful for cleaning and visualizing data
library(tidyverse)Next, I will import the statements, which are stored in a csv file. I will import and then store these statements in an object called “stats_raw” (i.e. raw statements). (Note I have already done some pre-procesing, which I decided to spare you from here. For example, I already eliminated responses from anyone who did not respond to all six questions with both a true statement and a false statement. I also eliminated some empty columns and renamed columns to make their meaning clearer.)
[Note, that you can toggle the display of the code, by clicking the button labeled “hide”/“code” on the right, above each code block,]
# First, load in the statements
stats_raw <- read.csv(file = "statements_final.csv")First, let’s look at some very basic features of the data set, to get a better understanding of its basic structure and organization.
To begin, let’s look at the basic dimensions of the dataset – that is the number of rows and columns.
dim(stats_raw)## [1] 5004 13Above, we see that the dataset has 5004 rows, and 13 columnns.
Let’s examine what those columns are.
# See the names of the columns
colnames(stats_raw)## [1] "per_id" "stat_id" "q_num" "grd_truth" "statement"
## [6] "stat_method" "stat_type" "order_first" "sex" "age"
## [11] "race" "rand" "rand_order"As we see, we have the following columns
| Column Name | Content |
|---|---|
| per_id | unique identification number for each participant |
| stat_id | unique identification number for each statement |
| q_num | the question (of the six) to which this statement is a response |
| grd_truth | “ground truth”, a column indicating whether the statment is a lie or a truth |
| statement | this contains the entire text of the participants response |
| stat_method | for each lie, the participants written explanation for how they generated that lie (empty for truths) |
| stat_type | for each lie, the participant’s multiple choice response for whether the response is based more on fantasy or reality (empty for truth) |
| order_first | the order in which participants completed the statements (lie followed by truths, or truths followed by lies) |
| sex | the participants self-reported sex |
| age | the participants self-reported age in years |
| race | the participants self-reported race/ethnicity |
| rand | a random number (previously used to sort the statements) |
| rand_order | a number ranking each statement by the previous random number (i.e. lowest random # gets 1, second lowest gets 2, etc) |
Finally, let’s take a little preview at a few full rows of the data, to also get a better sense of the structure. (Note: the columns appear in the order “per_id”, “stat_id”, “q_num”, etc. You can view more columns, by clicking the right pointing arrow following at the right of the right-most column name.)
head(stats_raw)Okay, now let’s look at some of the responses provided by participants, to each of the six questions.
The first question instructed participants to “Describe how you met [person you know well].” Let’s look at some of the responses participants gave.
Here is a sample of 5 truthful responses. (Note: I’ve re-arranged the columns here, so that the statement text appears as the firsct column.)
# display 5 true responses to question 1, by pre-computed random number
stats_raw %>%
filter(q_num == 1,
grd_truth == "truth") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderAnd here is a sample of 5 lies.
# display 5 untruthful responses to question 1, by pre-computed random number
stats_raw %>%
filter(q_num == 1,
grd_truth == "lie") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderThe second question asked participants: “What is something in your life that you regret (i.e. wish you had done differently)?”
Here is a sample of 5 truthful responses.
# display 5 true responses to question 2, by pre-computed random number
stats_raw %>%
filter(q_num == 2,
grd_truth == "truth") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderAnd here is a sample of 5 lies.
# display 5 untruthful responses to question 2, by pre-computed random number
stats_raw %>%
filter(q_num == 2,
grd_truth == "lie") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderThe third question instructed participants to: “Please describe what you did yesterday.”
Here is a sample of 5 truthful responses.
# display 5 true responses to question 3, by pre-computed random number
stats_raw %>%
filter(q_num == 3,
grd_truth == "truth") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderAnd here is a sample of 5 lies.
# display 5 untruthful responses to question 3, by pre-computed random number
stats_raw %>%
filter(q_num == 3,
grd_truth == "lie") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderThe fourth question instructed participants to: “Give some reason why you like [person you like/dislike]” (where the name of a person they listed as as person they like was piped in when they were asked for a truthful response, and the name of a person they listed as a person they disliked was piped in when they were asked for a un untruthful response).
Here is a sample of 5 truthful responses.
# display 5 true responses to question 4, by pre-computed random number
stats_raw %>%
filter(q_num == 4,
grd_truth == "truth") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderAnd here is a sample of 5 lies.
# display 5 untruthful responses to question 4, by pre-computed random number
stats_raw %>%
filter(q_num == 4,
grd_truth == "lie") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderThe fifth question instructed participants: “Describe a personal strength of yours (i.e. something that you are good at).”
Here is a sample of 5 truthful responses.
# display 5 true responses to question 5, by pre-computed random number
stats_raw %>%
filter(q_num == 5,
grd_truth == "truth") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderAnd here is a sample of 5 lies.
# display 5 untruthful responses to question 5, by pre-computed random number
stats_raw %>%
filter(q_num == 5,
grd_truth == "lie") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderIn the sixth question, participants were asked: “What is a hobby of yours or something that you enjoy doing in your free time? And why do you enjoy it?”
Here is a sample of 5 truthful responses.
# display 5 true responses to question 6, by pre-computed random number
stats_raw %>%
filter(q_num == 6,
grd_truth == "truth") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_orderAnd here is a sample of 5 lies.
# display 5 untruthful responses to question 6, by pre-computed random number
stats_raw %>%
filter(q_num == 6,
grd_truth == "lie") %>%
arrange(rand_order) %>%
select(statement,
everything()) %>%
top_n(5)## Selecting by rand_order