Nine Practical Guides to Support Your Research in 2026

In 2025, psychological scientists proposed a range of innovative developments in research practices across the field. The Observer has compiled a list of guides, tutorials, and manuals designed to support psychological scientists as they expand their toolboxes of research practices and methods. These papers were published in 2025 in the APS open-access journal Advances in Methods and Practices in Psychological Science. They are not listed in any particular order. 

See last year’s guides to support research practices.

1. A practical guide to specifying random effects in longitudinal dyadic multilevel models
Kareena S. del Rosario and Tessa V. West  

Analyzing over-time dyadic data can be challenging, particularly when using multilevel models with complex random-effect structures. This tutorial lays out the best practices of model specification for longitudinal dyadic multilevel modeling, providing a practical guide to specifying random effects with both theoretical and practical considerations in mind. The authors conclude with a discussion of alternative models and the utility of data simulation during study design, helping readers plan and select the best approach for their research. 

Resources: 

• Access the data and code for SAS and R.  

2. From embeddings to explainability: A tutorial on LLM-based text analysis for behavioral scientists
Rudolf Debelak, Timo K. Koch, Matthias Aßenmacher, and Clemens Stachl  

Large language models (LLMs) are transforming research in psychology and the behavioral sciences by enabling advanced text analysis at scale. This tutorial provides an accessible introduction to LLM-based text analysis, focusing on the Transformer architecture. The authors guide researchers through the process of preparing text data, using pretrained Transformer models to generate text embeddings, fine-tuning models for specific tasks such as text classification, and applying interpretability methods to explain model predictions. 

Resources: 

• Access the .csv files.  
• Read the supplemental material. 

3. The response-process-evaluation method: A new approach to survey item validation
Melissa G. Wolf, Elliott Ihm, Andrew Maul, and Ann Taves 

Pretesting survey items for interpretability and relevance is a commonly recommended practice in the social sciences. The goals are to construct items that are understood as intended by the population of interest and to test if participants use the expected cognitive processes when responding to a survey item. Existing methods of investigating item comprehension lack clear guidelines for retesting revised items and documenting improvements and can be difficult to implement in large samples. To remedy this, the authors introduce the response-process-evaluation method, a standardized framework for pretesting multiple versions of a survey. 

Resources: 

• See an example of a full evaluation report.
• Visit the appendix for more information.

4. Six fallacies in substituting large language models for human participants
Zhicheng Lin 

Can artificial-intelligence systems, such as large language models, replace human participants in behavioral and psychological research? This paper critically evaluates the replacement perspective and identifies six interpretive fallacies that undermine its validity. For each fallacy, specific safeguards are provided to guide responsible research practices. This framework provides a pathway for researchers to leverage language models productively while respecting the fundamental differences between machine intelligence and human thought. 

5. Open science in the developing world: A collection of practical guides for researchers in developing countries
Hu Chuan-Peng, Zhiqi Xu, Aleksandra Lazić, et al. 

Over the past decade, the open science movement has transformed the research landscape, although its impact has largely been confined to developed countries. Recently, researchers from developing countries have called for a redesign of open science to better align with their unique contexts. However, raising awareness alone is insufficient—practical actions are required to drive meaningful and inclusive change. This paper offers a four-level guide for gradual engagement and discusses potential pitfalls of current open science practices. 

Resources: 

• View guides for early career scientists in the supplemental material. 

6. Bridging null hypothesis testing and estimation: A practical guide to statistical conclusion drawing from research in psychology
Henk A. L. Kiers and Jorge N. Tendeiro 

A well-known problem of null hypothesis significance testing is that it cannot be used to find support for the null hypothesis. A common solution for this is to replace the exact 0 value by an interval associated with values that are close to 0. This approach is denoted as equivalence testing and is a special case of procedures that test intervals of values against each other. The researchers discuss three alternative general approaches, based on Bayesian analysis, that result in probabilities that can be interpreted as probabilities of the population parameters rather than probabilities of the data. The authors show how each of the methods works in the analysis of an example dataset and discuss their relative pros and cons. 

Resources: 

• Access the R script. 

Methods: How to Do Data Visualization Using R—Even If You Don’t Use R

7. Navigating unmeasured confounding in nonexperimental psychological research: A practical guide to computing and interpreting E-value
Kaiwen Bi, Gabriel J. Merrin, Tianyu Li, et al. 

Randomized experiments remain the gold standard for establishing causality, yet ethical and practical constraints in certain fields often require researchers to rely on observational data. In this tutorial, the authors explore the frequently overlooked but critical issue of unmeasured confounding in psychological research and introduce psychologists to the E-value, a novel and straightforward method for assessing the robustness of exposure–outcome associations to unmeasured confounding. They demonstrate the application of E-value using common psychological-research scenarios in R and discuss its strengths, limitations, and recommended best practices. 

Resources: 

• Access the R script. 
• Download the dataset. 

8. A primer on fixed effects and fixed-effects panel modeling using R, Stata, and SPSS
Nicolas Sommet and Oliver Lipps  

Fixed-effects modeling is a powerful tool for estimating within-clusters associations in cross-sectional data and within-participants associations in longitudinal data. Although commonly used by other social scientists, this tool remains largely unknown to psychologists. To address this issue, the authors offer a pedagogical primer tailored for this audience, complete with R, Stata, and SPSS scripts.  

Resources:

• Access the dataset and scripts for R, Stata, and SPSS.

9. An expert guide to planning experimental tasks for evidence-accumulation modeling
Russell J. Boag, Reilly J. Innes, Niek Stevenson, et al. 

Evidence-accumulation models (EAMs) are powerful tools for making sense of human and animal decision-making behavior. This article provides practical guidance for designing tasks appropriate for EAMs, relating experimental manipulations to EAM parameters, planning appropriate sample sizes, preparing data, and conducting an EAM analysis. Their advice is based on prior methodological studies and their substantial collective experience with EAMs. 

Feedback on this article? Email [email protected] or login to comment.