B.1 Positive psychology

B.1.1 Introduction

In a highly-cited publication, Seligman, Steen, Park, & Peterson (2005) found that several positive psychology interventions (PPIs) lastingly increased happiness and decreased depressive symptoms in a placebo-controlled internet study. The study by Woodworth et al. (2017) re-examined this claim by comparing the three most effective PPIs (identical with the interventions used by Seligman, Steen, Park, & Peterson, 2005) to a placebo control in a web-based, randomized assignment design. The corresponding data was published in a separate article (Woodworth, O’Brien-Malone, Diamond, & Schüz, 2018).

B.1.2 Data sources

Articles reporting original research:

• Seligman, M. E., Steen, T. A., Park, N., & Peterson, C. (2005). Positive psychology progress: Empirical validation of interventions. American Psychologist, 60(5), 410–421. doi: 10.1037/0003-066X.60.5.410

• Woodworth, R. J., O’Brien-Malone, A., Diamond, M. R., & Schüz, B. (2017). Web-based positive psychology interventions: A reexamination of effectiveness. Journal of Clinical Psychology, 73(3), 218–232. doi: 10.1002/jclp.22328

Article on data used here:

• Woodworth, R. J., O’Brien-Malone, A., Diamond, M. R. and Schüz, B. (2018). Data from, ‘Web-based Positive Psychology Interventions: A Reexamination of Effectiveness.’ Journal of Open Psychology Data, 6: 1. doi: 10.5334/jopd.35

• See https://openpsychologydata.metajnl.com/articles/10.5334/jopd.35/ for details.

• The dataset is available from figshare at https://doi.org/10.6084/m9.figshare.1577563.v1.

B.1.3 Codebook

Description of the variables and values contained in the 2 original data files:

B.1.3.1 File posPsy_participants.csv

The file posPsy_participants.csv contains 6 variables with demographic information on 295 participants:

1. id: Participant’s ID.

2. intervention: 3 positive psychology interventions (PPIs), plus 1 control condition:

   • 1 = “Using signature strengths,”
   • 2 = “Three good things,”
   • 3 = “Gratitude visit,”
   • 4 = “Recording early memories” (control condition).

3. sex:

   • 1 = female,
   • 2 = male.

4. age: Participant’s age (in years).

5. educ: Level of education:

   • 1 = Less than Year 12,
   • 2 = Year 12,
   • 3 = Vocational training,
   • 4 = Bachelor’s degree,
   • 5 = Postgraduate degree.

6. income:

   • 1 = below average,
   • 2 = average,
   • 3 = above average.

B.1.3.2 File posPsy_AHI_CESD.csv

The file posPsy_AHI_CESD.csv contains answers to the 24 items of the Authentic Happiness Inventory (AHI) and answers to the 20 items of the Center for Epidemiological Studies Depression (CES-D) scale (see Radloff, 1977) for multiple (1 to 6) measurement occasions:

1. id: Particpant ID.

2. occasion: Measurement occasion:

   • 0 = Pretest (i.e., at enrollment),
   • 1 = Posttest (i.e., 7 days after pretest),
   • 2 = 1-week follow-up, (i.e., 14 days after pretest, 7 days after posttest),
   • 3 = 1-month follow-up, (i.e., 38 days after pretest, 31 days after posttest),
   • 4 = 3-month follow-up, (i.e., 98 days after pretest, 91 days after posttest),
   • 5 = 6-month follow-up, (i.e., 189 days after pretest, 182 days after posttest).

3. elapsed.days: Time since enrolment measured in fractional days.

4. intervention: Intervention group (1 to 4).

5. ahi01–ahi24: Responses on 24 AHI items.

6. cesd01–cesd20: Responses on 20 CES-D items.

7. ahiTotal: Total AHI score.

8. cesdTotal: Total CES-D score.

B.1.4 Getting the data

Files available

The following files were generated from the original data files (and saved in .csv format):

1. posPsy_participants.csv: Original participant data (295 x 6 variables):
   http://rpository.com/ds4psy/data/posPsy_participants.csv.

2. posPsy_AHI_CESD.csv: Original data of dependent measures in long format (992 x 50 variables):
   http://rpository.com/ds4psy/data/posPsy_AHI_CESD.csv.

3. posPsy_AHI_CESD_corrected.csv: Corrected version of dependent measures in long format (990 x 50 variables):
   http://rpository.com/ds4psy/data/posPsy_AHI_CESD_corrected.csv.

4. posPsy_data_wide.csv: Corrected version of all data joined in wide format (295 x 294 variables):
   http://rpository.com/ds4psy/data/posPsy_data_wide.csv.
   Different measurement occasions are suffixed by .0, .1, …, .5.

Loading data

We can load data stored in csv-format into R by using the read_csv() function (from the readr package, which is part of the tidyverse). Here, we obtain the data files from online sources (at http://rpository.com/ds4psy/):

# Packages:
library(readr)

# Load csv-data files from online links:

# 1. Participant data: 
posPsy_p_info <- read_csv(file = "http://rpository.com/ds4psy/data/posPsy_participants.csv")
dim(posPsy_p_info)  # 295 x 6 
#> [1] 295   6

# 2. Original DVs in long format:
posPsy_AHI_CESD <- read_csv(file = "http://rpository.com/ds4psy/data/posPsy_AHI_CESD.csv")
dim(posPsy_AHI_CESD)  # 992 x 50
#> [1] 992  50

# 3. Corrected DVs in long format:
posPsy_long <- read_csv(file = "http://rpository.com/ds4psy/data/posPsy_AHI_CESD_corrected.csv")
dim(posPsy_long)  # 990 x 50
#> [1] 990  50

# 4. Transformed and corrected version of all data (in wide format): 
posPsy_wide <- read_csv(file = "http://rpository.com/ds4psy/data/posPsy_data_wide.csv")
dim(posPsy_wide)  # 295 x 294 
#> [1] 295 294

# Check number of missing values: 
sum(is.na(posPsy_p_info))   #     0 missing values 
#> [1] 0
sum(is.na(posPsy_AHI_CESD)) #     0 missing values 
#> [1] 0
sum(is.na(posPsy_long))     #     0 missing values 
#> [1] 0
sum(is.na(posPsy_wide))     # 37440 missing values!  
#> [1] 37440

B.1.5 References

Articles

• Radloff, L. S. (1977). The CES-D scale: A self report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. doi: 10.1177/014662167700100306

• Seligman, M. E., Steen, T. A., Park, N., & Peterson, C. (2005). Positive psychology progress: Empirical validation of interventions. American Psychologist, 60(5), 410–421.

• Woodworth, R. J., O’Brien-Malone, A., Diamond, M. R., & Schüz, B. (2017). Web-based positive psychology interventions: A reexamination of effectiveness. Journal of Clinical Psychology, 73(3), 218–232. doi: 10.1002/jclp.22328

• Woodworth, R. J., O’Brien-Malone, A., Diamond, M. R. and Schüz, B. (2018). Data from, ‘Web-based positive psychology interventions: A reexamination of effectiveness.’ Journal of Open Psychology Data, 6: 1. doi: 10.5334/jopd.35

Data

• See the files at https://doi.org/10.6084/m9.figshare.1577563.v1.

Note that it is non-trivial to read and re-format the provided data into the formats used above. Actually, doing so involves many steps and decisions and thus would make an excellent data science project (see Appendix C for details).

References

Seligman, M. E., Steen, T. A., Park, N., & Peterson, C. (2005). Positive psychology progress: Empirical validation of interventions. American Psychologist, 60(5), 410. https://doi.org/10.1037/0003-066X.60.5.410

Woodworth, R. J., O’Brien-Malone, A., Diamond, M. R., & Schüz, B. (2017). Web-based positive psychology interventions: A reexamination of effectiveness. Journal of Clinical Psychology, 73(3), 218–232. https://doi.org/10.1002/jclp.22328

Woodworth, R. J., O’Brien-Malone, A., Diamond, M. R., & Schüz, B. (2018). Data from ‘Web-based positive psychology interventions: A reexamination of effectiveness’. Journal of Open Psychology Data, 6(1). https://doi.org/10.5334/jopd.35