Linkoping University

Most of the structural problems examiners flag at public research universities show up in predictable places. A common one: an introduction that describes a broad topic rather than a specific, answerable question. Another: a literature review that summarizes sources one by one instead of showing where they disagree or where a gap remains — which then leaves the methodology chapter with nothing clear to justify. Results sections sometimes drift into interpretation that belongs in the discussion instead, and conclusions occasionally introduce new claims that were never actually argued in the body of the thesis. Knowing the twelve-part shape (title page, abstract, acknowledgements, contents, introduction, literature review, methodology, results, discussion, conclusion, references, appendices) is only useful once each part is doing the specific job it's meant to do, not just occupying the expected slot.

Momentum matters more than any single writing technique. Working in short, regular sessions with a concrete, measurable goal — a specific subsection drafted, one table analyzed, three sources summarized — tends to produce more usable pages over a semester than waiting for a large uninterrupted block of time that rarely arrives on schedule.

It also helps to write the methodology chapter close to when those decisions were actually made, since the reasoning behind a sampling choice or an analysis method is far easier to explain while it's fresh than to reconstruct from old notes weeks later.

Program type has a real effect on what a thesis at this level typically involves. Based on the programs Linkoping University is generally known for — Computer science, engineering, management — students here are likely working within one of a few common shapes: technical and data-heavy programs usually push methodology and results chapters toward statistical or computational analysis, while applied or professional programs (business, management, public-facing fields) more often lean on case-study framing, stakeholder analysis or applied recommendations instead of raw data collection.

Either way, the general principle holds: methodology should match what the research question actually needs, not what looks most impressive on paper.