Conducting and Visualizing Specification Curve Analyses

The goal of specr is to facilitate specification curve analyses (Simonsohn, Simmons & Nelson, 2019; also known as multiverse analyses, see Steegen, Tuerlinckx, Gelman & Vanpaemel, 2016). It can be used to investigate how different (theoretically plausible) analytical choices affect outcome statistics within the universe of one single data set.

It provides functions to setup, run, evaluate, and plot the multiverse of specifications. A simple usage example is provided below. For more information about the various functions and specific use cases, visit the documentation.

There are also some vignettes that exemplify and explain specific aspects and functions of the package:

Disclaimer

We do see a lot of value in investigating how analytical choices affect a statistical outcome of interest. However, we strongly caution against using specr as a tool to somehow arrive at a better estimate. Running a specification curve analysis does not make your findings any more reliable, valid or generalizable than a single analysis. The method is only meant to inform about the effects of analytical choices on results, and not a better way to estimate a correlation or effect.

Installation

Install specr from CRAN:

Or install the most recent development version from GitHub with:

# install.packages("devtools")
devtools::install_github("masurp/specr")

Usage

Using specr is comparatively simple. The main function is run_specs() in which analytical choices are specified as arguments. The function plot_specs() can then be used to visualize the results.

# Load package
library(specr)

# Run specs
results <- run_specs(df = example_data, 
                     y = c("y1", "y2"), 
                     x = c("x1", "x2"), 
                     model = c("lm"), 
                     controls = c("c1", "c2"), 
                     subsets = list(group1 = unique(example_data$group1),
                                    group2 = unique(example_data$group2)))
# Result frame
head(results)
#> # A tibble: 6 × 23
#>   x     y     model controls estimate std.error statistic  p.value conf.low
#>   <chr> <chr> <chr> <chr>       <dbl>     <dbl>     <dbl>    <dbl>    <dbl>
#> 1 x1    y1    lm    c1 + c2     4.95      0.525     9.43  3.11e-18    3.92 
#> 2 x2    y1    lm    c1 + c2     6.83      0.321    21.3   1.20e-57    6.20 
#> 3 x1    y2    lm    c1 + c2    -0.227     0.373    -0.607 5.44e- 1   -0.961
#> 4 x2    y2    lm    c1 + c2     0.985     0.324     3.04  2.62e- 3    0.347
#> 5 x1    y1    lm    c1          5.53      0.794     6.97  2.95e-11    3.96 
#> 6 x2    y1    lm    c1          8.07      0.557    14.5   6.90e-35    6.98 
#> # … with 14 more variables: conf.high <dbl>, fit_r.squared <dbl>,
#> #   fit_adj.r.squared <dbl>, fit_sigma <dbl>, fit_statistic <dbl>,
#> #   fit_p.value <dbl>, fit_df <dbl>, fit_logLik <dbl>, fit_AIC <dbl>,
#> #   fit_BIC <dbl>, fit_deviance <dbl>, fit_df.residual <int>, fit_nobs <int>,
#> #   subsets <chr>

# Plot
plot_specs(results, choices = c("x", "y", "controls", "subsets"))

How to cite this package

citation("specr")
#> 
#> To cite 'specr' in publications use:
#> 
#>   Masur, Philipp K. & Scharkow, M. (2020). specr: Conducting and
#>   Visualizing Specification Curve Analyses. Available from
#>   https://CRAN.R-project.org/package=specr.
#> 
#> A BibTeX entry for LaTeX users is
#> 
#>   @Misc{,
#>     title = {specr: Conducting and Visualizing Specification Curve Analyses (Version 0.2.2)},
#>     author = {Philipp K. Masur and Michael Scharkow},
#>     year = {2020},
#>     url = {https://CRAN.R-project.org/package=specr},
#>   }

References

Papers that used ‘specr’

If you have published a paper in which you used specr and you would like to be included in the following list, please send an email to Philipp.

  • Ballou, N., & van Rooij, A. J. (2021). The relationship between mental well-being and dysregulated gaming: a specification curve analysis of core and peripheral criteria in five gaming disorder scales. The Royal Society Open Science. https://doi.org/10.1098/rsos.201385

  • Ballou, N., & Zendle, D. (2022). “Clinically significant distress” in internet gaming disorder: An individual participant meta-analysis. Computers in Human Behavior, 129. https://doi.org/10.1016/j.chb.2021.107140

  • Burton, J.W., Cruz, N. & Hahn, U. (2021). Reconsidering evidence of moral contagion in online social networks. Nature Human Behaviour. https://doi-org.vu-nl.idm.oclc.org/10.1038/s41562-021-01133-5

  • Cosme, D., & Lopez, R. B. (2020, March 7). Neural indicators of food cue reactivity, regulation, and valuation and their associations with body composition and daily eating behavior. https://doi.org/10.1093/scan/nsaa155

  • Del Giudice, M., & Gangestad, S. W. (2021). A Traveler’s Guide to the Multiverse: Promises, Pitfalls, and a Framework for the Evaluation of Analytic Decisions. Advances in Methods and Practices in Psychological Science. https://doi.org/10.1177/2515245920954925

  • Henson, P., Rodriguez-Villa, E., Torous, J. (2021). Investigating Associations Between Screen Time and Symptomatology in Individuals With Serious Mental Illness: Longitudinal Observational Study Journal of Medical Internet Research, 23(3), e23144. https://doi.org/10.2196/23144

  • Kritzler, S., & Luhmann, M. (2021, March 25). Be Yourself and Behave Appropriately: Exploring Associations Between Incongruent Personality States and Positive Affect, Tiredness, and Cognitive Performance. https://doi.org/10.31234/osf.io/9utyj

  • Masur, P. K. (2021). Understanding the Effects of Conceptual and Analytical Choices on ‘Finding’ the Privacy Paradox: A Specification Curve Analysis of Large-Scale Survey Data. Information, Communication & Society. https://doi.org/10.1080/1369118X.2021.1963460