Quantitative Data Analysis

Sequence, 4 sessions, 5 days

A crucial step along the PhD journey is the processing, analysis, and interpretation of the data that students have collected. These data could be:

• Quantitative, where value is measured by use of numbers, or
• Qualitative, usually semi-structured data or textual data from interviews, or
• Mixed methods – a combination of qualitative and quantitative.

This sequence of sessions develops students’ skills in the use of quantitative analysis software. Specifically, students will learn:

• Cleaning and preparation of data for analysis.
• Data manipulation, including the creation of new variables by recoding and mathematical computation.
• Data summarisation.
• Bivariate analyses for quantitative and variables outcomes.
• Multivariable analyses for quantitative outcomes.
• Interpretation of commonly reported estimates, significance test results including confidence intervals.

Schedule the sequence on Qualitative Data Analysis immediately before or after this sequence. Your students will need those skills if they are to be successful researchers and teachers.

Outcomes
By the end of this sequence of sessions, students can:

• Prepare do-files.
• Clean data and prepare the dataset for analysis in line with the objectives.
• Create new variables and modify existing ones.
• Run bivariate and multivariable analyses and interpret results.

Preparation
Engage specialist co-facilitator/s and resource persons to support students.
Consult and share resources:

• StataCorp. 2021. Stata 18 Base Reference Manual. College Station, TX: Stata Press.
• Stata 18 documentation.

Well before these sessions:

• Provide STATA software for each student.
• Require all students to ensure that their installed STATA software is functional.
• Send three practice datasets to students, one each for analysis of quantitative data and longitudinal data.
• Send students an introduction to quantitative data analysis and a document with sample commands for the training sessions.
• Develop or find practice exercises for each session. Make sure that you and the resource persons test all the exercises.

Approach
For each session, each day, follow these guidelines:

• Use the STATA software throughout. Skills learnt on STATA should enable students to use other software.
• Use a practical hands-on approach to introduce the basic statistical concepts. Project or share your own screen to demonstrate the techniques involved (in data cleaning in preparation for analysis, data manipulation to create new variables, data exploration and summarisation, and significance testing). Also demonstrate the key manipulation procedures for longitudinal data.
• After your demonstration, students run the commands as you instruct them to. Remind them to ask questions if they encounter any challenges.
• Progress slowly through the sessions to enable slow learners to follow. Remind students that they can copy commands from the document and paste them in the STATA command window.
• Give students a practice exercise at the end of each session.
• At the end of each day of the training, give assignments to the students. Assess students’ logfiles each day and give feedback.

Assessment

• Assess and give feedback on students’ practice exercises with their data, based on topics covered during each session.
• Students must submit logfiles (and tables constructed from STATA output in some cases) for all assignments. Check the logfiles for errors in use of commands and application of principles.
• Students submit their do-files for assessment.
• If necessary – for example, if you run this sequence intensively over a single week – allow students additional time to submit exercises for assessment and feedback.

Steps