ilog3 Education Quantitative Study Design
Goal
Create a defensible quantitative research design aligned with research questions and variables.
Inputs
- Research questions/hypotheses
- Population and setting
- Available data or planned data collection
- Variable model from
education-variable-identification - Desired design: survey, experiment, quasi-experiment, regression, SEM, multilevel, longitudinal
Workflow
- Confirm design type and feasibility.
- Define population, sampling, inclusion/exclusion, and sample size logic.
- Finalize IV/DV/mediator/moderator/control variables.
- Identify instruments/scales/tests and reliability/validity evidence.
- Specify data collection procedure.
- Specify statistical analysis plan.
- Produce a method-section skeleton.
Tool Calls
Sample size/power:
G*Power: https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower
Open-source analysis:
install.packages(c("tidyverse", "psych", "car", "lavaan", "lme4", "effectsize", "mediation"))
No-code/low-code:
jamovi: https://www.jamovi.org/
JASP: https://jasp-stats.org/
Output Format
| Component | Design Decision | Rationale | Tool/Analysis |
|---|
Include:
- Hypotheses
- Variable table
- Sampling plan
- Instrument table
- Statistical analysis plan
- Method section draft
Quality Rules
- Causal claims require experimental/quasi-experimental logic.
- For pre/post studies, include baseline equivalence and appropriate controls.
- For nested education data, consider multilevel modeling.