# Project Components > * Brief descriptions of the stuff inside each component > * Download .html documents in [analyses](https://osf.io/ptx4f/) component for quick access to main analyses and supplemental descriptives and analyses, including plots > * Key supplement documents (see [analyses](https://osf.io/ptx4f/)) > * Sensitivity Analyses Shiny Web application ([https://nickmichalak.shinyapps.io/sounds_of_sickness_sensitivity/](https://nickmichalak.shinyapps.io/sounds_of_sickness_sensitivity/)) ## Supplemental Analyses Notes > We summarize key supplemental analyses in the [Supplementary Analyses Summaries](https://osf.io/4c7vr/wiki/Supplementary%20Analyses%20Summaries/) Project Wiki. In the analysis .html documents linked below, we encourage interested readers to inspect 3 key patterns, some of which are not discussed in the primary manuscript due to word limits: > 1. Accuracy depended on the interaction between Sound Origin (Infectious vs. non-Infectious) and Sound Type (coughs, sneezes, or sniffs). > 2. Accuracy did not consistently correlate with Germ Aversion nor Perceived Infectablity, which are subscales of the Perceived Vulnerability to Disease Questionnaire. > 3. In some cases, accuracy depended on Subjective Certainty. > In addition, we evaluated whether our null findings might be explained by demographic confounders and wrote up our study in a supplemental wiki: [Do target demographics suppress sound origin identification accuracy?](https://osf.io/4c7vr/wiki/Do%20target%20demographics%20suppress%20sound%20origin%20identification%20accuracy%3F/) > The links below will take you to a page where you can download a .html document, which you can view on any web browser. In the bullet points below, we tell you where to find key elements of the analysis in the floating table of contents , which "floats" (scrolls with you) on the left side of each .html. > * **Study 1 (Pilot Study)** > * [2study1_data_describe.html](https://osf.io/5c9st/) > * [3study1_data_analysis.html](https://osf.io/5hys3/) > * To inspect the Sound Origin x Sound Type interaction, see models >> model 3 >> interaction effect plot. > * To inspect the Sound Origin x PVDQ interaction, see models >> model 4 >> interaction effect plot. > * To inspect the Sound Origin x Certainty interaction, see models >> model 6 >> interaction effect plot. > * **Study 2** > * [2study2_data_describe.html](https://osf.io/z5rmx/) > * [3study2_data_analysis.html](https://osf.io/p3wny/) > * To inspect the Sound Origin x Sound Type interaction, see models >> model 3 >> interaction effect plot. > * To inspect the Sound Origin x PVDQ interaction, see models >> model 4 >> interaction effect plot. > * To inspect the Sound Origin x Certainty interaction, see models >> model 6 >> interaction effect plot. > * **Study 3** > * [2study3_data_describe.html](https://osf.io/fz7a4/) > * [3study3_data_analysis.html](https://osf.io/n5bfu/) > * To inspect the Sound Origin x Sound Type interaction, see models >> model 3 >> interaction effect plot. > * To inspect the Sound Origin x PVDQ interaction, see models >> model 4 >> interaction effect plot. > * To inspect the Sound Origin x Certainty interaction, see models >> model 6 >> interaction effect plot. > * **Study 4** > * [2study4_data_describe.html](https://osf.io/ew5g9/) > * [3study4_data_analysis.html](https://osf.io/atkqs/) > * To inspect the Sound Origin x Sound Type x Rating Condition interaction, see models >> model 3 >> interaction effect plot. > * To inspect the Sound Origin x Sound Type x Rating Condition x Sound Rating interaction, see models >> model 5 >> interaction effect plot. > * To inspect the Sound Origin x Rating Condition x PVDQ interaction, see models >> model 8 >> interaction effect plot. > * To inspect the Sound Origin x Rating Condition x Certainty interaction, see models >> model 10 >> interaction effect plot. > * **Combined Studies** > * [4study_all_data_plot.html](https://osf.io/my76d/) > * We used this code to create the plots in the paper (also some other plots which can be viewed in the figures subfolder in [analyses](https://osf.io/ptx4f/)). > * [4exploratory_meta_analysis.html](https://osf.io/2yza9/) > * In these analyses we summarize via an exploratory internal meta-analysis all four studies using two approaches: (1) a fixed effects meta-analysis and a Bayesian logistic mixed effects model. Specifically, we (1) meta-analyze the intercept and infectious vs. non-infectious sound origin logits, and (2) we combine the data from all studies to fit a similar (but Bayesian) mixed effects model. > * **Acoustic Feature Analysis** > * [4audioclip_analysis.html](https://osf.io/j5ztx/) > * We used the R package [soundgen](https://cran.r-project.org/web/packages/soundgen/index.html) to extract acoustic features from each sound clip, and then we evaluated (1) feature differences between infectious and non-infectious categories and (2) whether a logistic regression of sound features can accurately classify (non-)infectious category [analyses](https://osf.io/ptx4f/)). ## Materials > * surveys (.doc and .qsf) and images used in all studies > * can use .qsf to import survey into Qualtrics ([Importing a Survey](https://www.qualtrics.com/support/survey-platform/survey-module/survey-tools/import-and-export-surveys/#ImportingASurvey)) > * sound stimuli used in all studies (these are also embedded in .qsf files) ## Data > * We prepared datasets using the preparation scripts (see preparation scripts in [analyses](https://osf.io/ptx4f/) component) > * All "clean" data are deidentified (i.e., we do not share raw data, but we share the scripts used to generate cleaned data from raw data) > * Each data folder includes study-specific data > 1. **codebook.** variable names, Qualtrics labels, and some category labels for stimuli (bare bones) > 2. **qualtrics.** de-identified, prepared datasets in "wide form" (i.e., each participant has a unique row) > 3. **lqualtrics.** de-identified, prepared datasets in "long form" (i.e., each observation has a unique row) > 4. **turkprime.** turkprime HIT data ## Analyses > * Preparation, Description, and Analyses Scripts > 1. **preparation.** prepare and deidentify data for description and analyses > 2. **describe.** descriptives (including turkprime payment demographics), psychometrics, and correlations > 3. **analysis.** contrasts, raw data plots, models, model diagnostic plots, result summaries, and effect plots > 4. Combined analyses use data from multiple studies > * "2017F_sickness_sounds.Rproj" > * .Rproj refers to an R project file which automatically sets your R session's working directory to the folder that the .Rproj is in (for more details, see [Using Projects](https://support.rstudio.com/hc/en-us/articles/200526207-Using-Projects) or [Workflow: projects](https://r4ds.had.co.nz/workflow-projects.html)) > * For your own analyses, we recommend following these steps: > 1. download the analyses component > 2. unzip the analyses component on your computer > 3. download and save the data component in the unzipped analyses folder > 4. unzip the data component > 2. open the .Rproj (this opens R Studio and sets the working directory for you) ### **Analyses Document Naming Scheme** > e.g., "2study3_data_describe.html" > 1. The first number in filenames order the files top-to-bottom within the folder (1 = preparation, 2 = description, 3 = analysis, 4 = multiple-study analysis) > 2. Study number (Study 1-4 and not reported studies) > 3. Script type (preparation, description, analysis) > 4. Filetype (.Rmd or .html) > * .Rmd refers to an R Markdown script which can be mostly easily viewed and edited in R Studio (it can be read as plain text) > * .html includes the output from the .Rmd script and can be opened in any internet browser (e.g., Google Chrome, Safari) > * **example.** "2study3_data_describe.html" refers to the output of the descriptive statistics R Markdown script for Study 3 (e.g., demographics, correlations, means, standard deviations) ## Other Registration > * registration summaries (i.e., "need to know" details for evaluating protocol that was planned vs. reported-in-manuscript) are included as .csv and .xlsx files ## app > * Users can conduct simple sensitivity analyses via an R Shiny web application ([https://nickmichalak.shinyapps.io/sounds_of_sickness_sensitivity/](https://nickmichalak.shinyapps.io/sounds_of_sickness_sensitivity/)).