Understanding validity in psychological research

Validity is a key concept in psychological research that assesses whether the findings are genuine and meaningful. It focuses on two main aspects: whether the study measures what it intends to measure, and whether the results can be applied more broadly.

Internal validity

Internal validity refers to the extent to which a study demonstrates a true cause-and-effect relationship between variables, without interference from confounding factors.

External validity

External validity concerns how well the findings can be generalised beyond the specific research setting, including to other people, places, or times.

Internal validity and threats to it

Internal validity ensures that the results of a study reflect a true cause-and-effect relationship. However, various threats can compromise this, leading to inaccurate conclusions.

Key threats to internal validity

• Social desirability bias - This occurs when participants alter their responses or behaviour to appear more favourable, rather than reflecting their true thoughts or feelings.
• Subjectivity - Researchers' personal biases, opinions, or emotions can influence how they interpret participants' behaviour, leading to skewed results.
• Demand characteristics - These are cues in the research environment that signal to participants what the study is about, causing them to change their behaviour to match perceived expectations.
• Low controls - When extraneous variables (factors outside the main variables of interest) are not properly managed, they can affect the dependent variable (the outcome being measured).

Checking validity in a psychological study

Researchers often include specific steps to verify the validity of their measures. An example comes from Baron-Cohen et al.'s study on the 'eyes test', which assesses the ability to interpret mental states from the eye region of faces.

Method for checking validity

The researchers selected words and foils (distractor options) for each set of eyes. Two authors initially chose the words, and then an eight-member judging team evaluated them. For a word to be accepted as correct, at least five judges had to agree on it.

Results of the validity check

This process ensured that the words accurately represented the emotions depicted in the eyes, with high agreement among judges confirming the choices.

Evaluation of the approach

This method strengthened internal validity by reducing subjectivity in word selection.

External validity and its components

External validity determines how applicable research findings are outside the original study conditions. It includes several subtypes, each addressing different aspects of generalisation.

Generalisability and population validity

Generalisability refers to the extent to which research findings can be applied to people beyond those who participated in the study. This is also known as population validity, which focuses on whether the sample represents the target population (the broader group the research aims to understand).

Findings should only be generalised if the sample is representative, meaning it mirrors the key characteristics of the target population.

Factors affecting generalisability:

• Sampling technique - Random sampling tends to produce more representative samples, improving generalisability. In contrast, opportunity sampling or volunteer sampling can lead to biased samples, reducing population validity.

Ecological validity

Ecological validity measures how well research findings can be generalised from the study's artificial setting to real-life environments.

Characteristics of ecological validity:

• Real-life settings - Research in natural environments, like field experiments, often has higher ecological validity because behaviour is less influenced by artificial conditions.
• Natural behaviour - When studies allow participants to act as they would normally, generalisation is easier. However, lab-based studies with controlled but unrealistic settings can have low ecological validity.

Temporal validity

Temporal validity assesses whether research findings remain applicable across different time periods.

Influences on temporal validity:

• Social changes - Norms, values, and behaviours evolve over time, which can make older findings outdated. For example, Milgram's 1960s obedience study might have different results today due to shifts in societal attitudes towards authority.
• Susceptibility in social research - Areas like social psychology are particularly prone to temporal validity issues because human attitudes and interactions change with historical events, laws, or cultural trends.