Technical Directions
Intervention Analysis Tool - Technical Directions
​
This tool, called the Intervention Analysis Tool (IAT), allows users to compare intervention periods to non-intervention periods by analyzing data from a CSV or Excel file. Users can choose from three different designs (AB, ABA, ABAB), and the tool generates a summary table, effect size table, and two plots (color and black and white) to visualize the data. The IAT can be useful for researchers, educators, or other professionals who want to evaluate the effectiveness of interventions in a variety of settings.
​
About the Tool
The IAT is a Shiny Web Application developed by Matthew Courtney in 2023. It uses the R statistical programming language to read data off of a spreadsheet and create a summary of any column. The IAT is hosted on the shinyapps.io server.
​
Preparing your Data
-
Collect and organize your data: First, you need to collect the data that you want to use in the tool. Ensure that the data is relevant and suitable for the analysis you intend to perform. Then, organize the data into a suitable format, such as a CSV or Excel file.
-
Ensure data quality: Before using the data, you need to check its quality. Ensure that the data is complete, free from errors, and formatted correctly. You can use tools like OpenRefine or Excel to clean and validate the data.
-
Select the appropriate variables: Choose the variables that you want to use in the analysis. Selecting the right variables is essential to achieve accurate and meaningful results. Ensure that the variables you choose are relevant to the research question or analysis that you want to perform.
-
Prepare the data: Prepare the data for analysis by performing transformations such as scaling, normalization, or encoding. This will depend on the nature of the data and the analysis you want to perform. You can use tools like Pandas or Scikit-learn in Python to perform these transformations.
-
Split the data: Split the data into two parts: training data and testing data. The training data is used to build the model, while the testing data is used to evaluate the model's accuracy. The proportion of data split will depend on the size of your dataset and the type of analysis you want to perform.
-
Save the data: Finally, save the prepared data in a format that is compatible with the tool you intend to use. This will depend on the tool's requirements, so make sure to check the documentation for instructions on data format and input requirements.
​
Using the IAT
-
Upload your data: Start by uploading your prepared data to the tool, ensuring that it is in the correct file format (CSV, TSV, or Excel). You can do this by either dragging and dropping the file into the designated area or by clicking on the "Upload" button and navigating to the file on your computer.
-
Set the parameters: Once your data is uploaded, you will be asked to set the parameters for your analysis. These will vary depending on the specific tool you are using, but could include things like choosing the type of analysis to run, selecting the columns to analyze, setting thresholds or confidence intervals, and specifying the desired output format.
-
Run the analysis: After you have set the parameters, click on the "Run" button to start the analysis. Depending on the size of your data set and the complexity of the analysis, this may take some time to complete.
-
Interpret the output: Once the analysis is complete, the tool will generate an output file. This could be a table of results, a graph or chart, or a summary report. It is important to carefully review and interpret the output to understand what the analysis is telling you.
-
Refine and re-run as necessary: Based on the output, you may need to refine your analysis and re-run it with different parameters or settings. This iterative process is common in data analysis, and the tool you are using should allow you to easily adjust and rerun your analysis until you get the results you need.
​
When interpreting the output of your analysis, it is important to keep in mind the specific parameters you set and the limitations of the tool you are using. For example, if you set a confidence interval for a particular result, you should be aware that this means there is a certain degree of uncertainty in the outcome. It is also a good idea to consult additional resources and seek out expert advice as needed to ensure that you are properly interpreting your results and using them to make informed decisions.
​
References
-
R Core Team. (2020). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/
-
Winston Chang, Joe Cheng, JJ Allaire, Yihui Xie and Jonathan McPherson (2021). shiny: Web Application Framework for R. R package version 1.7.1. https://CRAN.R-project.org/package=shiny
-
Hadley Wickham (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. ISBN 978-3-319-24277-4, https://ggplot2.tidyverse.org.
-
Mark A. Lipsey, David B. Wilson (2001). Practical Meta-Analysis, SAGE Publications, Inc., ISBN: 9780761921673. https://cran.r-project.org/web/packages/effsize/index.html
-
Bavaresco, L. D., Scortegagna, S. A., Kuhn, T., Freitas, M. P., & Mello, R. O. (2020). SCVA: An R package for single-cell variability analysis. Bioinformatics, 36(12), 3876-3878. https://doi.org/10.1093/bioinformatics/btaa297
-
Hadley Wickham and Jennifer Bryan (2019). readxl: Read Excel Files. R package version 1.3.1. https://CRAN.R-project.org/package=readxl
-
Hadley Wickham, Romain François, Lionel Henry and Kirill Müller (2021). dplyr: A Grammar of Data Manipulation. R package version 1.0.7. https://CRAN.R-project.org/package=dplyr