How to learn new statistical techniques

In this section, I share resources for learning different statistical technics and concepts. These resources are introductory materials and should be treated as jumping off points for learning new techniques. Resources are grouped first by technique and then by type of resource.

Steps to learning a new statistical technique (and using it in your work)

  1. Find and read in detail one (or two) introduction paper explaining in laymen's word what the technique is, what it's useful for, the basic steps involved in carrying out the technique, and how results from this technique are interpreted.
  2. Find and read a WELL-written paper (preferably in your field) in which authors use that technique. Good, well-written papers using sophisticated techniques often (if not always) include information about what the technique is, why they chose to use it and how they will intrepret the results obtained from it. While some might argue that it is not the researchers' job to teach the reader these techniques, the author will inherently do so when justifying the use of the technique and elaborating their hypotheses (again, especially when the paper is well-written).
  3. Find a tutorial along with a practice dataset and try applying the analyses to that dataset.
  4. Finally, try applying the technique to one of your own datasets (or a dataset on which haven't been analyzed with this technique yet).

Statistics

Basic (Frequentist) Statistical Concepts

Basic Bayesian Statistical Concepts and Tests

Meta-analysis

Simulation

  • Article: Hallgren, Kevin A. “Conducting Simulation Studies in the R Programming Environment.” Tutorials in Quantitative Methods for Psychology 9, no. 2 (2013): 43–60.

Measurement

Item response theory (IRT)

Generalizability theory

  • Article (simpler): Lakes, Kimberley D., and William T. Hoyt. “Applications of Generalizability Theory to Clinical Child and Adolescent Psychology Research.” Journal of Clinical Child and Adolescent Psychology : The Official Journal for the Society of Clinical Child and Adolescent Psychology, American Psychological Association, Division 53 38, no. 1 (2009): 144–65. https://doi.org/10.1080/15374410802575461.
  • Article (more theoretical): Briesch, Amy M., Hariharan Swaminathan, Megan Welsh, and Sandra M. Chafouleas. “Generalizability Theory: A Practical Guide to Study Design, Implementation, and Interpretation.” Journal of School Psychology 52, no. 1 (February 1, 2014): 13–35. https://doi.org/10.1016/j.jsp.2013.11.008.

Modeling

Structural equation modeling

Structural equation modeling combines CFA with regression allowing one to model relationships (correlation or regression) between latent variables (error free-measures).

Advanced related topics: Bifactor models, Longitudinal SEM, Multilevel SEM

Multilevel modeling

Multilevel modeling consists of regression while accounting for dependencies in the dataset. It can be used to analyze data composed of individuals (level 1) belonging to clusters like classrooms (level 2) or mulitple data points (level 1) belonging to the same individual (level 2).

  • Book: Finch, W. H., Bolin, J. E., & Kelley. (2019). Multilevel Modeling Using R. Second Edition. New York: CRC Press.

Drift diffusion modeling

Signal detection theory

Neuroimaging

Functional Magnetic Resonance Imaging (fMRI)

  • Website course: https://www.newbi4fmri.com/
  • Book: Poldrack, R., Mumford, J., & Nichols, T. (2011). Handbook of Functional MRI Data Analysis. Cambridge: Cambridge University Press. doi:10.1017/CBO9780511895029

Multivoxel pattern analysis (MVPA)

Multivoxel pattern analysis refers to a group of statistical techniques in which the pattern of activation across multiple voxels is used to make inferences (e.g., based on decoding or similarity). This is in contrast to univariate analyses which assesses the amount of activation (high vs low activation) for a given region or voxel.

  • Article: Weaverdyck, Miriam E, Matthew D Lieberman, and Carolyn Parkinson. “Tools of the Trade Multivoxel Pattern Analysis in FMRI: A Practical Introduction for Social and Affective Neuroscientists.” Social Cognitive and Affective Neuroscience 15, no. 4 (April 1, 2020): 487–509. https://doi.org/10.1093/scan/nsaa057.