Research in Physical Education I: Importance and Classification, Ethical Issues, Descriptive, Historical and Experimental Methods and Experimental Designs, the Research Problem, Hypotheses and their Testing, Tools of Research, Literature Search and Critical Reading, and the Preparation of the Research Proposal, Report, Abstract and Papers

Historical research studies the past, of sport, the Olympic movement, physical education or an institution, from primary sources (original documents, records, eyewitness accounts, photographs, artefacts) and secondary sources (accounts by others). It evaluates its sources by external criticism, which establishes whether a source is genuine (its authenticity, date and authorship), and internal criticism, which establishes whether its content is accurate and credible. Descriptive research describes what exists: surveys (of fitness, participation or facilities), case studies of an individual, team or institution, developmental studies (longitudinal, following the same subjects over time, or cross-sectional, comparing different age groups at one time), correlational studies of relationships, normative studies that establish norms, and job and activity analyses. Experimental research manipulates an independent variable (such as a training programme) under controlled conditions to observe its effect on a dependent variable (such as VO2max), and is the only method that can establish cause and effect.

Experimental designs, as classified by Campbell and Stanley (1963), are pre-experimental (the one-shot case study, the one-group pretest–posttest design and the static-group comparison, all weak because they lack a proper control group or random assignment), true experimental (the pretest–posttest control group design, the posttest-only control group design and the Solomon four-group design, all with random assignment to groups) and quasi-experimental (the non-equivalent control group design and time-series designs, used when intact groups such as school classes must be used). Factorial designs study two or more independent variables at once, such as training type and sex. Internal validity is the confidence that the independent variable, and not something else, caused the change; it is threatened by history, maturation, testing (practice on the pretest), instrumentation, statistical regression, selection bias and experimental mortality (drop-out). External validity is the extent to which the results can be generalised to other people and settings.

A research problem is identified from personal experience in teaching and coaching, the literature and its recommendations for further research, theory, new developments in sport and technology, and social needs. It is formulated as a clear, specific statement, often as a question, that is researchable, significant, feasible in terms of time, cost, subjects and equipment, ethical, and within the researcher's competence. Delimitations are the boundaries the researcher deliberately sets (such as male university hockey players aged 18 to 25); limitations are weaknesses outside the researcher's control (such as the subjects' diet or motivation). Operational definitions state how each variable will be measured.

A hypothesis is a tentative, testable statement of the expected relationship between variables. The research hypothesis states what the researcher expects; the null hypothesis (H0) states that there is no difference or relationship, and it is the null hypothesis that statistical tests actually test; the alternative hypothesis (H1) states that there is. Hypotheses may be directional ("plyometric training will increase vertical jump more than weight training"), tested with a one-tailed test, or non-directional ("the two programmes will differ in their effect"), tested with a two-tailed test. A good hypothesis is testable, clear, consistent with known facts, specific and stated in advance. In testing, the researcher sets a level of significance, usually 0.05 or 0.01, which is the probability of a Type I error: rejecting a null hypothesis that is actually true. A Type II error is failing to reject a null hypothesis that is actually false; its probability is β, and the power of the test, 1 − β, is the probability of detecting a real effect, which rises with sample size and effect size. If the p-value obtained is below the chosen level, the null hypothesis is rejected and the result is called statistically significant.

A questionnaire is a set of written questions answered by respondents themselves; it may be closed (fixed choices, easy to analyse) or open (free answers, richer but harder to analyse), and may be mailed, handed out or online. It is economical for large samples but suffers from low return rates and cannot probe answers, so it is pilot-tested, kept short and clear and sent with a covering letter. An opinionnaire, or attitude scale, measures opinions and attitudes rather than facts, by rating scales such as the Likert scale (degrees of agreement, commonly five points) or the Thurstone scale (statements with pre-assigned values). An interview collects data face to face or by telephone; it may be structured (fixed questions in a fixed order), semi-structured or unstructured, gives high response rates and allows clarification and probing, but is time-consuming and open to interviewer bias. Observation records behaviour as it happens, as in a study of teacher behaviour or of game actions; it may be participant or non-participant and structured (with checklists, rating scales or systematic coding) or unstructured, and video recording now aids reliability.

The literature search locates what is already known, refines the problem, suggests methods and avoids duplication. Its sources are the library (books, encyclopaedias, handbooks, bound journals, theses), bibliographic and full-text databases (such as SPORTDiscus for sport, PubMed for biomedicine, ERIC for education, and Scopus and Web of Science for citation searching), India's Shodhganga repository of doctoral theses run by INFLIBNET, internet search engines and scholarly search tools such as Google Scholar, and online and open-access journals. Its steps are defining the topic and key words, searching with Boolean operators (AND narrows, OR broadens, NOT excludes) and filters, screening titles and abstracts, obtaining and reading the relevant sources, and recording full references. Note-taking should record the full citation, the purpose, method, sample, findings and conclusions of each study and the reader's own comments, distinguishing quotation from paraphrase to avoid plagiarism. Critical reading asks whether the design suited the question, whether the sample and measures were adequate, whether the statistics were appropriate, whether the conclusions follow from the results, and whether the source is peer-reviewed and current.

A research proposal (synopsis) sets out the plan of a study before it begins, usually as: title; introduction and background; statement of the problem; objectives; hypotheses; delimitations and limitations; definitions of terms; significance of the study; a brief review of related literature; methodology (design, subjects and sampling, variables and the tests and tools with their reliability, procedure and data collection, and statistical analysis); a time schedule and budget; and references. A research report or thesis follows the traditional chapters: introduction, review of related literature, methodology (procedure), analysis of data and results with discussion, and summary, conclusions and recommendations, followed by the bibliography and appendices.

An abstract is a brief, self-contained summary, commonly about 150–300 words, of the purpose, method, main results and conclusions, written last and placed first, with key words. A paper for publication in a journal usually follows the IMRaD structure (Introduction, Methods, Results and Discussion), is written to the journal's guidelines and style (such as the APA style, now in its seventh edition), and passes through peer review before acceptance. A paper for presentation at a conference is shorter and planned for the time allotted: a clear structure, a few key messages, readable slides with figures rather than dense tables, rehearsal to time, and preparation for questions; a poster presentation displays the same content visually. Care should be taken to publish in reputable, peer-reviewed journals and to avoid predatory journals that publish without genuine review.