Personality Tests
Take This
First things first, please go and take the Big Five Personality Test.
Read Up
Then read this short synopsis on the test from FiveThirtyEight.
Data Files
Please download the files we’ll need.
Prerequisites
Open up RStudio and create a new script by going to File > New File > R Script. Save this in an easily accessible folder. Now unzip this week’s data set and take the files - Big5data.csv, Big5codebook.csv, Big5measures.csv, and 03-walkthrough - and drop them all in a folder of their own. If you aren’t familiar with how directories and file paths work on a computer, consider keeping both your R script and data sets in the same folder. It just makes life easier.
Before we load the libraries, we’re going to grab a package called surveytools2 that is not on the Comprehensive R Archive Network (CRAN) - the main R repository of packages. In any case, there is a command within the surveytools2 package that will make our lives easier moving forward. To install it, please run the following commands
install.packages("remotes")
remotes::install_github("peterhurford/surveytools2")
If you are asked to update any packages, please select 1: All.
Now please go ahead and load up the following libraries or download and load if needed
library("tidyverse")
library("psych")
##
## Attaching package: 'psych'
## The following object is masked from 'package:fontawesome':
##
## fa
## The following objects are masked from 'package:ggplot2':
##
## %+%, alpha
library("corrr")
library("surveytools2")
##
## Attaching package: 'surveytools2'
## The following object is masked from 'package:base':
##
## replace
Then set the working directory to the location of the script by running
setwd(dirname(rstudioapi::getActiveDocumentContext()$path))
in your console or by selecting Session > Set Working Directory > To Source File Location.
Data Files
We’ll be looking at a very truncated data set from the 2018 BigFive Personality Test. The full data set is a huge file with 1 million answers to the 50 personality items! You can access it here but be warned that opening a file that size can take a long time or may even crash some computers depending on their processing power and RAM. If you are interested in or confused by the formatting, it is a tab separated csv file.
Side Note: Loading the Entire Data Set
If you want to give the entire data set a go, in RStudio you can use
toobig_five_data <- read_delim("data-final.csv", delim="\t")
to save it to the big_five_data variable which you can access by running
toobig_five_data
At this point you probably want to just open up the csv files in Excel and if that is what works, then by all means. While I try to run everything in R, sometimes its just easier to double click it. The problem really lies in the size of a data set. The larger the data set, the more power and time it takes for excel to load it up. With R you avoid this whole idea and the set loads up within seconds, rather than minutes. If you try loading the original data set in both R and Excel, the difference is very apparent.
In any case, we will be working with data strictly in R, but never feel bad about opening and wrangling a data set using another application like Excel. For all of my snarkiness towards anything Microsoft, Excel has some nice qualities but don’t tell anyone I ever wrote that.
Loading a Local Data Set
Let’s turn back to the data set we’ll be using. To load the data set, codebook, and measures, run the following
big_five_data <- read_csv("Big5data.csv")
## Rows: 100 Columns: 50
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## dbl (50): EXT1, EXT2, EXT3, EXT4, EXT5, EXT6, EXT7, EXT8, EXT9, EXT10, EST1,...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
big_five_codebook <- read_csv("Big5codebook.csv")
## Rows: 50 Columns: 2
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (2): ID, Item
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
big_five_measures <- read_csv("Big5measures.csv")
## Rows: 5 Columns: 2
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): Description
## dbl (1): Measure
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
and take a look at each
big_five_data %>%
head()
## # A tibble: 6 × 50
## EXT1 EXT2 EXT3 EXT4 EXT5 EXT6 EXT7 EXT8 EXT9 EXT10 EST1 EST2 EST3
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 3 2 3 4 4 2 3 3 4 2 4 3 5
## 2 1 5 1 5 1 4 1 5 4 5 4 2 4
## 3 2 5 4 3 1 5 1 3 4 5 2 5 5
## 4 2 3 3 3 3 2 2 4 2 3 5 1 5
## 5 3 4 3 3 3 3 3 2 3 4 3 5 3
## 6 3 1 5 1 4 1 5 4 4 2 5 2 5
## # … with 37 more variables: EST4 <dbl>, EST5 <dbl>, EST6 <dbl>, EST7 <dbl>,
## # EST8 <dbl>, EST9 <dbl>, EST10 <dbl>, AGR1 <dbl>, AGR2 <dbl>, AGR3 <dbl>,
## # AGR4 <dbl>, AGR5 <dbl>, AGR6 <dbl>, AGR7 <dbl>, AGR8 <dbl>, AGR9 <dbl>,
## # AGR10 <dbl>, CSN1 <dbl>, CSN2 <dbl>, CSN3 <dbl>, CSN4 <dbl>, CSN5 <dbl>,
## # CSN6 <dbl>, CSN7 <dbl>, CSN8 <dbl>, CSN9 <dbl>, CSN10 <dbl>, OPN1 <dbl>,
## # OPN2 <dbl>, OPN3 <dbl>, OPN4 <dbl>, OPN5 <dbl>, OPN6 <dbl>, OPN7 <dbl>,
## # OPN8 <dbl>, OPN9 <dbl>, OPN10 <dbl>
with respondents self-report under the five factors Extroversion (EXT), Agreeableness (AGR), Conscientiousness (CSN), Emotional Stability (EST), and Intellect/Imagination (OPN).
big_five_codebook
## # A tibble: 50 × 2
## ID Item
## <chr> <chr>
## 1 EXT1 I am the life of the party.
## 2 EXT2 I don't talk a lot.
## 3 EXT3 I feel comfortable around people.
## 4 EXT4 I keep in the background.
## 5 EXT5 I start conversations.
## 6 EXT6 I have little to say.
## 7 EXT7 I talk to a lot of different people at parties.
## 8 EXT8 I don't like to draw attention to myself.
## 9 EXT9 I don't mind being the center of attention.
## 10 EXT10 I am quiet around strangers.
## # … with 40 more rows
big_five_measures
## # A tibble: 5 × 2
## Measure Description
## <dbl> <chr>
## 1 5 Agree
## 2 4 Somewhat Agree
## 3 3 Neutral
## 4 2 Somewhat Disagree
## 5 1 Disagree
For this walkthrough, let’s just concentrate on the Extroversion items which are given by EXT. That means we should probably pair down both the data set and codebook. For data sets with multiple columns, we can use the select() syntax. Moreover since we are looking at all of the columns that have EXT, we can pull only those using the starts_with() command. For more examples of options that can be used with select(), take a look here.
big_five_data_EXT <-
big_five_data %>%
select(starts_with("EXT"))
and then we’ll do something similar to the codebook. While we can’t simply select columns because all of the variables are in a single column. Much like in Excel, we can filter rows. To do this in R, we use the filter() command.
big_five_codebook_EXT <-
big_five_codebook %>%
filter(str_detect(ID, "EXT"))
big_five_data_EXT
## # A tibble: 100 × 10
## EXT1 EXT2 EXT3 EXT4 EXT5 EXT6 EXT7 EXT8 EXT9 EXT10
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 3 2 3 4 4 2 3 3 4 2
## 2 1 5 1 5 1 4 1 5 4 5
## 3 2 5 4 3 1 5 1 3 4 5
## 4 2 3 3 3 3 2 2 4 2 3
## 5 3 4 3 3 3 3 3 2 3 4
## 6 3 1 5 1 4 1 5 4 4 2
## 7 3 3 4 2 5 2 5 2 4 3
## 8 3 4 1 3 0 4 1 4 2 5
## 9 2 4 2 4 4 1 3 1 2 5
## 10 3 4 4 2 3 2 3 5 1 4
## # … with 90 more rows
big_five_codebook_EXT
## # A tibble: 10 × 2
## ID Item
## <chr> <chr>
## 1 EXT1 I am the life of the party.
## 2 EXT2 I don't talk a lot.
## 3 EXT3 I feel comfortable around people.
## 4 EXT4 I keep in the background.
## 5 EXT5 I start conversations.
## 6 EXT6 I have little to say.
## 7 EXT7 I talk to a lot of different people at parties.
## 8 EXT8 I don't like to draw attention to myself.
## 9 EXT9 I don't mind being the center of attention.
## 10 EXT10 I am quiet around strangers.
Method: Reverse Coding
Sometimes items can be worded ‘backwards’. In these situation, scales need to be reversed to point in the same direction.
Example
For example both these questions might be used to measure the same underlying construct or idea.
- I really like being alone
- I really don’t enjoy large gatherings
In this situation, if we used the same response scale for each participants’ responses they would simply cancel out.
When aligning the direction, we typically reverse code the one that has a negative, aka the negatively-keyed item. In this case is the second example since it includes the word don’t. There are exceptions to this rule, but they are narrowly defined. For example, someone who states that they “really don’t enjoy large gatherings” doesn’t automatically imply they like being alone. In fact, there are arguably an infinite number of possibilities of things they do like! The only time you can be code all items in either direction are in circumstances where there is a binary choice (e.g. Yes/No questions).
But wait…there’s more! You also have to consider the intent of the questions and the participants. This second and final check is crucial to reporting reliable and consistent results and is the primary reason why many surveys are simply not good.
Getting back to the main data set, let’s again look at the codebook.
big_five_codebook_EXT
## # A tibble: 10 × 2
## ID Item
## <chr> <chr>
## 1 EXT1 I am the life of the party.
## 2 EXT2 I don't talk a lot.
## 3 EXT3 I feel comfortable around people.
## 4 EXT4 I keep in the background.
## 5 EXT5 I start conversations.
## 6 EXT6 I have little to say.
## 7 EXT7 I talk to a lot of different people at parties.
## 8 EXT8 I don't like to draw attention to myself.
## 9 EXT9 I don't mind being the center of attention.
## 10 EXT10 I am quiet around strangers.
Well it certainly looks like EXT2, EXT4, EXT6, EXT8 and EXT10 are all negative in the context of of assessing extroversion. Furthermore while EXT9 includes the term don’t, it doesn’t orient the statement in a negative way so it stays the way it is.
Remember that command we needed from the surveytools2 package? Well it was to reverse code in a tidy way
big_five_data_EXT_rev <-
big_five_data_EXT %>%
mutate(EXT2 = reverse_code(EXT2),
EXT4 = reverse_code(EXT4),
EXT6 = reverse_code(EXT6),
EXT8 = reverse_code(EXT8),
EXT10 = reverse_code(EXT10))
big_five_data_EXT_rev
## # A tibble: 100 × 10
## EXT1 EXT2 EXT3 EXT4 EXT5 EXT6 EXT7 EXT8 EXT9 EXT10
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 3 4 3 1 4 3 3 2 4 3
## 2 1 1 1 0 1 1 1 0 4 0
## 3 2 1 4 2 1 0 1 2 4 0
## 4 2 3 3 2 3 3 2 1 2 2
## 5 3 2 3 2 3 2 3 3 3 1
## 6 3 5 5 4 4 4 5 1 4 3
## 7 3 3 4 3 5 3 5 3 4 2
## 8 3 2 1 2 0 1 1 1 2 0
## 9 2 2 2 1 4 4 3 4 2 0
## 10 3 2 4 3 3 3 3 0 1 1
## # … with 90 more rows
Let’s take a look at a sample of the first six rows between the two tables to see if the columns were indeed reverse coded.
Original Data Set | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| EXT1 | EXT2 | EXT3 | EXT4 | EXT5 | EXT6 | EXT7 | EXT8 | EXT9 | EXT10 |
| 3 | 2 | 3 | 4 | 4 | 2 | 3 | 3 | 4 | 2 |
| 1 | 5 | 1 | 5 | 1 | 4 | 1 | 5 | 4 | 5 |
| 2 | 5 | 4 | 3 | 1 | 5 | 1 | 3 | 4 | 5 |
| 2 | 3 | 3 | 3 | 3 | 2 | 2 | 4 | 2 | 3 |
| 3 | 4 | 3 | 3 | 3 | 3 | 3 | 2 | 3 | 4 |
| 3 | 1 | 5 | 1 | 4 | 1 | 5 | 4 | 4 | 2 |
Data Set with Reverse Coded Columns | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| EXT1 | EXT2 | EXT3 | EXT4 | EXT5 | EXT6 | EXT7 | EXT8 | EXT9 | EXT10 |
| 3 | 4 | 3 | 1 | 4 | 3 | 3 | 2 | 4 | 3 |
| 1 | 1 | 1 | 0 | 1 | 1 | 1 | 0 | 4 | 0 |
| 2 | 1 | 4 | 2 | 1 | 0 | 1 | 2 | 4 | 0 |
| 2 | 3 | 3 | 2 | 3 | 3 | 2 | 1 | 2 | 2 |
| 3 | 2 | 3 | 2 | 3 | 2 | 3 | 3 | 3 | 1 |
| 3 | 5 | 5 | 4 | 4 | 4 | 5 | 1 | 4 | 3 |
That looks correct! Notice that in the green columns, the numbers flip along the neutral measure (3), or maybe a better observation is that the numerical complement is correct if adding two corresponding cells equals 5. If you have had any upper level mathematics courses, you may notice these as a modulo or mod 5 system.
Computing the Internal Consistency
There are many approaches when assessing reliability, namely those associated with interviews, experiments, surveys, trials, etc. Internal consistency is foremost one of many existing methods used to measure reliability, and it is typically useful in assessing items (e.g from a protocol, questionnaire, test, etc). It essentially estimates to what degree total scores on an instrument would change or vary if different items were used. Here are four but really three approaches used by a majority of practitioners:
Composite Reliability
Getting a bit ahead of ourselves, composite reliability is based on the factor lodgings in a confirmatory factor analysis (CFA). We’ll cover this soon!
Cronbach’s alpha
By far the most utilized approach, it is also very procedural which is typically viewed as beneficial by many evaluators. There are many ways to tell R to compute alpha ($\alpha$) but by far the simplest may be to use the psych package
psych::alpha(big_five_data_EXT_rev)
##
## Reliability analysis
## Call: psych::alpha(x = big_five_data_EXT_rev)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.9 0.9 0.91 0.48 9.1 0.015 2.7 0.95 0.49
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.87 0.9 0.93
## Duhachek 0.87 0.9 0.93
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## EXT1 0.89 0.90 0.91 0.49 8.5 0.016 0.0123 0.50
## EXT2 0.89 0.89 0.90 0.47 8.1 0.017 0.0115 0.49
## EXT3 0.89 0.89 0.90 0.47 8.0 0.017 0.0111 0.49
## EXT4 0.89 0.89 0.90 0.46 7.7 0.017 0.0126 0.45
## EXT5 0.89 0.89 0.89 0.46 7.8 0.017 0.0098 0.46
## EXT6 0.89 0.89 0.90 0.48 8.4 0.016 0.0104 0.49
## EXT7 0.89 0.89 0.90 0.47 7.9 0.017 0.0119 0.46
## EXT8 0.90 0.90 0.91 0.50 9.1 0.015 0.0092 0.50
## EXT9 0.89 0.89 0.90 0.48 8.3 0.016 0.0128 0.50
## EXT10 0.89 0.89 0.90 0.47 7.8 0.017 0.0120 0.47
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## EXT1 100 0.67 0.67 0.61 0.59 2.8 1.2
## EXT2 100 0.73 0.74 0.71 0.66 3.4 1.2
## EXT3 100 0.75 0.75 0.72 0.68 3.2 1.3
## EXT4 100 0.79 0.79 0.78 0.73 2.0 1.2
## EXT5 100 0.79 0.79 0.78 0.73 3.3 1.3
## EXT6 100 0.68 0.68 0.65 0.60 2.7 1.3
## EXT7 100 0.78 0.77 0.74 0.70 2.9 1.5
## EXT8 100 0.59 0.59 0.53 0.49 1.8 1.3
## EXT9 100 0.71 0.71 0.67 0.63 3.3 1.3
## EXT10 100 0.78 0.78 0.75 0.71 1.4 1.3
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 miss
## EXT1 0.01 0.18 0.19 0.34 0.21 0.07 0
## EXT2 0.00 0.09 0.13 0.30 0.26 0.22 0
## EXT3 0.00 0.14 0.17 0.17 0.34 0.18 0
## EXT4 0.12 0.24 0.24 0.28 0.11 0.01 0
## EXT5 0.02 0.11 0.13 0.20 0.38 0.16 0
## EXT6 0.08 0.14 0.13 0.35 0.29 0.01 0
## EXT7 0.01 0.24 0.13 0.27 0.16 0.19 0
## EXT8 0.24 0.21 0.21 0.24 0.09 0.01 0
## EXT9 0.01 0.11 0.20 0.15 0.33 0.20 0
## EXT10 0.33 0.23 0.21 0.15 0.07 0.01 0
There is a lot there and most of those outcomes are useful, but for reliability we really care about the raw_alpha, or the the standardized alpha based upon the correlations. For a raw $\alpha$ score, the following breakdown is widely accepted but not without controversy.
| Cronbach’s Alpha | Measure of Internal Consistency |
|---|---|
| `\alpha\geq0.90` | Excellent |
| `0.70 >\alpha\geq0.90` | Good |
| `0.80 >\alpha\geq0.70` | Acceptable |
| `0.70 >\alpha\geq0.60` | Poor |
| `\alpha < 0.50` | Unacceptable |
Note that the thresholds themselves do not have any logical grounding, rather what is an acceptable threshold may be situation dependent.
If you’d like to access the alpha value easily, implement this
psych::alpha(big_five_data_EXT_rev)$total$raw_alpha
## [1] 0.9002546
which for a quick and dirty measure of reliability seems to be pretty Good!
Inter-item Correlation
This is foremost an average which uses correlations.
big_five_data_EXT_rev %>%
correlate()
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
## # A tibble: 10 × 11
## term EXT1 EXT2 EXT3 EXT4 EXT5 EXT6 EXT7 EXT8 EXT9 EXT10
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EXT1 NA 0.363 0.443 0.494 0.454 0.289 0.546 0.363 0.478 0.435
## 2 EXT2 0.363 NA 0.415 0.500 0.644 0.628 0.506 0.315 0.438 0.543
## 3 EXT3 0.443 0.415 NA 0.577 0.631 0.379 0.566 0.295 0.496 0.629
## 4 EXT4 0.494 0.500 0.577 NA 0.471 0.595 0.540 0.501 0.452 0.636
## 5 EXT5 0.454 0.644 0.631 0.471 NA 0.495 0.649 0.246 0.582 0.551
## 6 EXT6 0.289 0.628 0.379 0.595 0.495 NA 0.356 0.373 0.315 0.544
## 7 EXT7 0.546 0.506 0.566 0.540 0.649 0.356 NA 0.359 0.559 0.509
## 8 EXT8 0.363 0.315 0.295 0.501 0.246 0.373 0.359 NA 0.450 0.396
## 9 EXT9 0.478 0.438 0.496 0.452 0.582 0.315 0.559 0.450 NA 0.391
## 10 EXT10 0.435 0.543 0.629 0.636 0.551 0.544 0.509 0.396 0.391 NA
The package gives an NA for variables that try to correlate with themselves. By default each variable should be perfectly correlated with itself but who cares? With the diagonal owned by NAs, we can obtain the average correlation of each item with all others by computing the means for each column EXT1 - EXT10.
inter_measure <-
big_five_data_EXT_rev %>%
correlate() %>%
select(-term) %>%
colMeans(na.rm = TRUE)
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
inter_measure
## EXT1 EXT2 EXT3 EXT4 EXT5 EXT6 EXT7 EXT8
## 0.4295075 0.4834744 0.4922734 0.5295312 0.5249037 0.4414833 0.5100294 0.3665847
## EXT9 EXT10
## 0.4623084 0.5150927
We can check the variables that are more strongly correlated with the other items in the EXT group. The same information is presented by arranging categories and descending internal consistency measures, respectively.
big_five_correlations_cat <-
data.frame(inter_measure) %>%
rownames_to_column() %>%
as_tibble() %>%
arrange(desc(inter_measure))
big_five_correlations_num <-
big_five_correlations_cat %>%
left_join(big_five_codebook, c("rowname" = "ID"))
big_five_correlations_cat
## # A tibble: 10 × 2
## rowname inter_measure
## <chr> <dbl>
## 1 EXT4 0.530
## 2 EXT5 0.525
## 3 EXT10 0.515
## 4 EXT7 0.510
## 5 EXT3 0.492
## 6 EXT2 0.483
## 7 EXT9 0.462
## 8 EXT6 0.441
## 9 EXT1 0.430
## 10 EXT8 0.367
big_five_correlations_num
## # A tibble: 10 × 3
## rowname inter_measure Item
## <chr> <dbl> <chr>
## 1 EXT4 0.530 I keep in the background.
## 2 EXT5 0.525 I start conversations.
## 3 EXT10 0.515 I am quiet around strangers.
## 4 EXT7 0.510 I talk to a lot of different people at parties.
## 5 EXT3 0.492 I feel comfortable around people.
## 6 EXT2 0.483 I don't talk a lot.
## 7 EXT9 0.462 I don't mind being the center of attention.
## 8 EXT6 0.441 I have little to say.
## 9 EXT1 0.430 I am the life of the party.
## 10 EXT8 0.367 I don't like to draw attention to myself.
Side Note: Getting a Jump on Joins
Next week we’ll get more into depth about joins, namely the left_join() but for now if you can’t wrap your head around what happened in that step, that is absolutely fine! If you want to get a basic rundown on joins, take a look at
STAT545.
In any case, it looks like EXT4, EXT5, EXT10, and EXT7 have the strongest correlation. The list below has been arranged to represent the data in descending order of measure.
| Variable | Internal Consistency Measure | Item |
|---|---|---|
| EXT4 | 0.5295312 | I keep in the background. |
| EXT5 | 0.5249037 | I start conversations. |
| EXT10 | 0.5150927 | I am quiet around strangers. |
| EXT7 | 0.5100294 | I talk to a lot of different people at parties. |
| EXT3 | 0.4922734 | I feel comfortable around people. |
| EXT2 | 0.4834744 | I don’t talk a lot. |
| EXT9 | 0.4623084 | I don’t mind being the center of attention. |
| EXT6 | 0.4414833 | I have little to say. |
| EXT1 | 0.4295075 | I am the life of the party. |
| EXT8 | 0.3665847 | I don’t like to draw attention to myself. |
and the overall average inter-item correlation can be found by
mean(inter_measure)
## [1] 0.4755189
Item-total Correlation
The item-total correlation is similar to the correlation described previously. We are calculating the average but this time, we’re doing it by row rather than by column. To do that we sum
agg_scores <-
big_five_data_EXT_rev %>%
mutate(score = rowMeans(select(.,1:10))) %>%
correlate() %>%
focus(score) %>%
arrange(desc(score))
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
agg_scores
## # A tibble: 10 × 2
## term score
## <chr> <dbl>
## 1 EXT4 0.791
## 2 EXT5 0.789
## 3 EXT7 0.776
## 4 EXT10 0.775
## 5 EXT3 0.749
## 6 EXT2 0.732
## 7 EXT9 0.714
## 8 EXT6 0.680
## 9 EXT1 0.666
## 10 EXT8 0.594
Let’s Stop Here
Well that is quite enough for now. There are other ways to assess reliability. As you move along in the course, keep in mind that none of these could yield good measures for internal consistency. You may have to look around the internet for other R commands and packages that may be a better fit. As usual, you always have your peers and me if you get stuck!
Looking Ahead
Next week we get into uncovering underlying traits and groupings that a data set may have using an EFA. While there certainly a level of complexity that has to be understood, the return in information you can get about a measurable data set far outweighs the cost.