Other Research Strategies: Secondary Data, Experiments, Case Studies, and Reviews
❦
Secondary and Archival Data
Sources in Sri Lanka: annual reports of listed companies (Colombo Stock Exchange); Central Bank of Sri Lanka data; Department of Census and Statistics; Export Development Board and Board of Investment publications; industry association reports; international databases (World Bank, IMF); and commercial databases if your library subscribes.
Steps
1. Define the variables and how they are measured (for example, return on assets, leverage, firm size).
2. Define the sample (for example, listed manufacturing firms from 2015 to 2024) and inclusion criteria.
3. Collect data and record the source of each variable.
4. Clean the data: check for errors, outliers, and missing values; adjust for inflation if relevant.
5. Choose models (Chapter 11) and test assumptions.
6. Interpret and check robustness.
Data availability, changes in accounting standards, and the effects of the economic crisis can affect comparability, so describe these limitations.
Experiments and Quasi-Experiments
An experiment manipulates an independent variable (for example, two advertisements) and assigns participants randomly to conditions, then compares outcomes (Shadish et al., 2002). Quasi-experiments lack random assignment, and so need careful controls (for example, a pre-test and post-test with a comparison group). Report manipulation checks, sample sizes per group, and effect sizes. Online experiments and vignette studies are practical in business research.
Case Study Research
A case study investigates a contemporary phenomenon in depth within its real context, using several sources (interviews, documents, observation) (Yin, 2018). Design elements: research question, propositions, unit of analysis, case selection (single or multiple), data collection protocol, analysis (pattern matching, explanation building, cross-case synthesis), and criteria for quality. Eisenhardt (1989) describes theory-building from cases.
Delphi and Design Science
The Delphi method gathers expert opinions through several anonymous rounds until consensus emerges; it is used to develop frameworks and forecasts. Design science research builds and evaluates an artefact (a model, tool, or method) to solve a problem.
Systematic Literature Reviews and Bibliometric Studies
A systematic review follows a protocol: a focused question, explicit search strategy, inclusion and exclusion criteria, screening, quality appraisal, and synthesis, reported with a flow diagram (Page et al., 2021; Tranfield et al., 2003). Bibliometric analysis uses publication and citation data (for example, with Scopus or Web of Science exports and software such as VOSviewer) to map the intellectual structure of a field (Donthu et al., 2021). These can form the main method of a master's or doctoral study if rigorously done.
Action Research
A cycle of planning, acting, observing, and reflecting with participants to improve practice in an organization; common in management and education; clarify your role and manage bias.
CHAPTER 15