Research in Physical Education Questions and Answers

Answer: C

The most similar group is ordinal, interval and ratio data.

All three of these scales have order: values can be ranked from less to more. Interval and ratio data also have equal intervals, and ratio data a true zero, but all three share the property of order.

Nominal data only names categories (such as gender or sport), with no order at all. Each of the other options mixes nominal data with ordered scales, so its members are less alike.

Answer: C

The relationship between two ordinal variables is assessed with Spearman's rho.

Spearman's rank-order correlation works on ranks, so it is the appropriate correlation coefficient when both variables are ordinal, for example the ranks of athletes in two competitions, or when interval data are not normally distributed.

Cramer's V and phi measure association between nominal variables (phi for 2 × 2 tables), and the chi-square test tests independence of categorical variables. None of them uses the order in ordinal data.

Answer: A

The first stage would be to read introductory textbooks.

Before searching for research articles, the researcher needs a general understanding of the topic: its main concepts, terms, theories and key authors. Introductory textbooks provide that background and help in choosing the keywords and the scope for the actual search.

An online database search, asking a teacher for articles and ordering inter-library loans all come later, once the researcher knows what to look for.

Answer: B

Maturation and mortality are threats to internal validity.

Internal validity is the degree to which changes in the dependent variable can be attributed to the treatment rather than to other causes. In a longitudinal study, participants grow and change naturally over time (maturation), and some drop out or are lost (experimental mortality, attrition), so the remaining group may differ from the original. Both offer alternative explanations for any change observed.

Significance is a statistical decision, instrumentation is itself another threat to internal validity, and reliability concerns the consistency of measurement.

Answer: B

The correct matching is (1)-(ii), (2)-(iii), (3)-(iv), (4)-(i).

• Developmental research studies changes in behaviour across the years of life (1-ii).
• Longitudinal research follows the same participants over a period of years (2-iii).
• Cross-sectional research compares samples from different age groups at one time to assess the effects of maturation (3-iv).
• Correlational research explores relationships among variables (4-i).

The other codes swap the longitudinal and cross-sectional designs or misplace correlational research; the two developmental designs are the easiest to confuse.

Answer: B

A constant takes a fixed value.

In research, a constant is a characteristic that has the same value for every subject in a study, for example age when all the participants are 15, or the distance of a test run. A variable, by contrast, takes different values from one subject to another.

A constant is not necessarily zero, and its value is not 'nothing'; it is simply fixed at one value. 'Not fixed' describes a variable, so it is the opposite of the answer.

Answer: A

The commands are in the Data menu.

In SPSS, the Data menu holds commands that act on the whole data file and its cases, including Sort Cases, Merge Files, Split File (to run analyses separately for groups) and Select Cases (to analyse a subset).

The Transform menu creates and recodes variables, for example Compute Variable and Recode, and the Analyze menu runs statistical procedures such as t-tests and regression. There is no 'Variable' menu in SPSS.

Answer: D

The research form is phenomenology.

Phenomenological research seeks to understand how individuals experience a phenomenon, for example what it is like for athletes to return from serious injury. The researcher collects detailed accounts, usually through in-depth interviews, and describes the essence of the lived experience.

Grounded theory aims to generate a theory from data. Ethnography studies the culture of a group, usually through long participant observation. A case study examines a single case, such as a team or programme, in depth, rather than the meaning of an experience.

Answer: D

The sample is a convenience sample.

A convenience sample is made up of the participants who happen to be easily available to the researcher. Here the schools taking part are simply those working on Saturday, the day the researcher chose to conduct the study; they were not selected by any chance procedure, so the sample is one of convenience and may not represent all urban schools.

A random sample gives every member of the population an equal chance of selection, a stratified sample draws from defined subgroups, and a cluster sample randomly selects whole groups such as schools.

Answer: D

Multistage cluster sampling helps in all these situations, so the answer is All of the above.

In multistage cluster sampling the researcher randomly selects large clusters, such as districts, then smaller clusters within them, such as schools, and finally individuals.

This is useful when the population is widely dispersed geographically, because data collection is concentrated in a few selected areas, which saves time and money on travel. And because selection is random at every stage, it remains a probability sample whose results can be generalised to the population. Each statement is a genuine advantage.