Descriptive statistics

Qualitative data

data expressed in words → collected from an interview or an unstructured observation

Strengths:

• offers a broader richness of detail than quantitative data.

• therefore is able to gain more insight into participants answers and achieve more external validity.

Limitations:

• Difficult to analyse as the data can not be summarised statistically which can be used to be identify patterns and make comparisons.

• It can be hard to draw conclusions as they are reliant on inferences which is subject to bias.

Quantitative data

expressed numerically → collected in the form of individual scores from participants from a study

Strengths:

• simple to analyse and this means that conclusions and comparisons can be drawn easier, without need for investigator bias.

Limitations:

• Can lack detail and may fail to represent ‘real life’ due to the oversimplification of answers.

Primary data (or field research)

original data collected specifically for purpose of investigation by the researcher.
collected first-hand from participants themselves.
→ usually from an experiment, questionnaire, interview or observation

Strengths:

• Authentic data is collected for the specific experiment and data collection methods can be designed specifically to target information the researcher requires.

Limitations:

• Requires time and effort for the researcher.

• Experiments can be costly and require planning, preparation and resources, while secondary data can be accessed in minutes.

Secondary data (or desk research)

data collected by someone other than the person who is conducting research
→ journal articles, books or websites

Strengths:

• Cheap and easily accessible with minimum effort.

• There is often no need to gather primary data if secondary data is already available.

Limitations:

• Variation in quality and accuracy of secondary data could effect validity of conclusions.

• Data could be outdated or incomplete.

Meta-analysis

Uses secondary data
A number of studies with the same aims/ hypothesis have their results pooled together and a joint conclusion is made.

Strengths:

• can create a larger, more varied sample where results can be generalised across larger populations → increasing external validity

Limitations:

• Prone to publication bias → May not select all relevant studies and choose to leave out those with negative or non-significant results, so conclusions of meta-analyses can be biased.

Interval data

exact measurements e.g. time taken to run a mile; anything measured in objective public units such as height or speed: scores on a test if test items are standardised/ of equal difficulty.
Each participant’s individual score can be identified.

Ordinal data

position/ rank e.g. who came 1^st, 2^nd and 3^rd on a test.
or any data that is measured on a rating scale (since its subjective)
or items on a test if the items are not of equal difficulty/ the test is not standardised.

Nominal data

we measure how many people fall in each category e.g. how many people passed or failed a test.
You can’t identify an individuals score.