Session, 90 minutes
Before researchers began data collection, they developed a data analysis plan to guide them from the initial stages of summarising and describing the data through to testing the hypotheses. At this point – as they return from the field with the data to answering their research question – guide them to revisit and refine their data analysis plan for quantitative, qualitative, and mixed-method studies.
Research questions are often framed broadly and need to be clarified and funnelled down into testable hypotheses and action steps. Having a clear plan of action is also important for research integrity and quality, as it guards against data-driven results and allows analyses to be reproduced.
Your aim, as the facilitator, is to give practical support to students to put their research thoughts into a plan of action in order to meet the objectives of their studies.
By the end of this sequence of sessions, students can:
- Discuss the essentials of a good data analysis plan.
- Identify the ingredients of a good data analysis plan.
- Create an appropriate data analysis plan for a quantitative, qualitative or a multi-method study.
- Generate dummy tables for quantitative data analysis based on specific study objectives.
For you, as the facilitator
Prepare guidance (Step 1).
Ensure that students prepare or revise their data analysis plan and dummy tables (for quantitative analysis) before the session.
|15 minutes||1. Present (quantitative) data analysis planning||Facilitator, full group|
|45 minutes||2. Peer review data analysis plans||Small groups|
|30 minutes||3. Revise data analysis plans and tables||Individual students|
Step 1: Present (quantitative) data analysis planning
In your presentation, remind students of the essence of a good data analysis plan:
- A plan of action.
- An investigator’s guide.
Note the key elements in creating a data analysis plan:
- Research questions or objectives (what to examine).
- Study design (how the questions will be addressed or examined).
- Data sources, study population – specifying the inclusion/exclusion criteria.
- Study measures: detailed definitions and derivations (including categorisation used, if any).
- Research instruments (tools and questions to be used).
- Definition of variables in terms of:
- Main exposure variables.
- Outcome variable(s) and independent variables.
- Level of measurements – nominal-, ordinal-, interval-, and ratio-level variables.
- Levels of analysis (univariate, bivariable, and multivariable analysis).
- Level of acceptable significance.
- Proper tests.
Note other details to consider including in a data analysis plan:
- Other covariates, including potential confounders and mediators.
- Sub-groups: does the main effect vary by sub-groups of participants?
- Missing data and methods for dealing with missing data (such as coding missing values as separate categories, imputation methods).
- Sequence of planned analyses, including:
- Statistical methods.
- How hypotheses will be tested.
- How potential confounders and biases will be assessed and addressed.
- Planned tables and figures, dummy tables.
Step 2: Peer review data analysis plans
In groups of four, students review each other’s data analysis plans.
Step 3: Revise data analysis plans and tables
30 minutes and after the session
Individual students revise their data analysis plans and tables in light of guidance and peer reviews. They submit their revised plans for you or co-facilitators to review.