Primary and secondary data, and alternative hypotheses

Primary and secondary data

• Primary data are collected by the researcher for their own study, for example from their questionnaire, interview or experiment. They fit the aim exactly and the researcher controls how they were gathered, but they take time and money.
• Secondary data already exist and were collected by someone else, for example published studies, official statistics or another researcher's data set. They are quick and cheap and may cover large samples, but may not fit the aim and their quality cannot be checked.

A meta-analysis such as Bond and Smith (1996) uses secondary data.

Alternative hypotheses

A hypothesis is a precise, testable prediction. In non-experimental studies (questionnaires, correlations) it is called an alternative hypothesis; in experiments, an experimental hypothesis.

• Non-directional (two-tailed): predicts a difference but not which way. 'There will be a difference in self-reported obedience scores between males and females.'
• Directional (one-tailed): predicts the direction. 'Females will report higher obedience scores than males.'

The null hypothesis predicts no difference (any difference is due to chance).

Writing a good hypothesis

Name both groups or conditions and operationalise the measure: say exactly how it is scored, for example 'total score on a 10-item Likert obedience questionnaire (10 to 50)'. Use a directional hypothesis when previous research suggests the direction; otherwise use non-directional.

Worked example

Write a directional alternative hypothesis for a questionnaire comparing conformity in students who play team sports and those who do not.

1. Name both groups: team sport players and non-players.
2. Operationalise conformity: score on a scenario-based conformity questionnaire.
3. State the direction.

Answer: Students who play team sports will score higher on a scenario-based conformity questionnaire than students who do not play team sports.

Check yourself

1. A researcher uses government crime statistics to study obedience to the law. These are
   • primary data
   • secondary data
   • qualitative data only
   • an alternative hypothesis
   Show answer

   secondary data. The data were collected by someone else for another purpose.

2. Which is a non-directional alternative hypothesis?
   • Older adults will have higher conformity scores than younger adults
   • Younger adults will have higher conformity scores than older adults
   • There will be no difference in conformity scores between older and younger adults
   • There will be a difference in conformity scores between older and younger adults
   Show answer

   There will be a difference in conformity scores between older and younger adults. It predicts a difference without saying which way.

3. What is one advantage of primary data over secondary data?
   • It is always cheaper
   • The researcher knows exactly how it was collected and it fits the aim
   • It always has a larger sample
   • It never needs ethical approval
   Show answer

   The researcher knows exactly how it was collected and it fits the aim. Control over collection and fit to the aim.

4. What does a null hypothesis predict?
   • A difference in a stated direction
   • A difference in an unknown direction
   • No difference, apart from chance
   • A correlation of +1
   Show answer

   No difference, apart from chance. The null predicts no real difference or relationship.

5. Rewrite this hypothesis so that it is operationalised: 'Men are more obedient than women.'
   Show answer

   Men will score higher than women on total self-reported obedience, measured by a 10-item Likert questionnaire.. Say how obedience is measured, and keep both groups and the direction.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification (issue January 2026)
• Pearson International A Level Psychology component guide: WPS01
• IAL Psychology revision site: Self-reports

Last updated 8 November 2026. Spotted a mistake? Tell us.