Model Papers

Below we list some papers that give examples, information or guidelines on or can act as models (i.e. you can follow the same method and structure for your specific situation and often get a high-quality result) for different types of studies that are typical in software engineering. They are organised based on the basic type of question to be answered. Note! If the papers/PDFs are not directly linked below you need to go through your library to get a copy; do NOT ask me for a copy.

Guideline papers are marked "G", papers that are examples of using a particular method are marked "E", and model papers are marked "M" (can be models for whole studies and ways to report on a certain type of study).

1. Will a new/changed method/tool improve our software development?

G: Case studies for method and tool evaluation (Kitchenham, Pickard, Pfleeger)
G: CASE tool evaluation: experiences from an empirical study (Budgen, Thomson)
G: An empirical incremental approach to tool evaluation and improvement (Miller, MacDonald)

2. How to study a particular case/company?

G: Guidelines for conducting and reporting case study research in SE (Runeson, Höst)

3. How to report on the context for an empirical study in industry?

G: Context in Industrial Software Engineering Research (Petersen, Wohlin) - template for how to describe the industrial context in case studies.

4. What is the state-of-the-art in a certain field?

G/M: Systematic literature reviews in SE - A systematic literature review (Kitchenham et al)
M: Regression Testing Minimisation, Selection and Prioritisation : A Survey (Yoo, Harman) - Good example that a thorough survey need not be based on a systematic search, excellent summary of state-of-the-art, issues and future of field.

5. How to analyse and report on interview studies?

M: The effect of moving from a plan-driven to an incremental software development approach with agile practices (Petersen, Wohlin) - 33 interviews on Ericsson.

6. How to rank/prioritize issues/solutions elicited from interview studies?

E: The effect of moving from a plan-driven to an incremental software development approach with agile practices (Petersen, Wohlin) - two types of ranking used in this paper. "Commonality" counts by how many different interviewees a certain issue was mentioned ("General" >= 1/3rd of interviewees, "Very common" >= 1/5th, "Common" >= 1/10th, "Other" = mentioned by at least one). But paper also used the "Top-three-count" method by explicitly asking for the top three issues in the interviews and then ranked them on how many times they were mentioned.