Measuring concepts

Definition (#f7aeae)

Important (#edcae9)

Extra (#fffe9d)

Measuring behavior:

Behaviour can be measured in many ways, a common measurement is to ask people how they feel.

Validity:

  • Refers to truth or accuracy and is supported by available evidence.

  • Construct Validity: Whether the measure that is used actually measures the construct it is intended to measure.

  • The validity of a psychological tool is determined by whether it measure this construct.

Indicators of Construct Validity:

  1. Face Validity: Involves only judgment of whether the content of the measure appears to actually measure the variable.

  2. Content Validity: Comparing the content of the measure with content that defines the construct.

  3. Concurrent Validity: To examines the relationship between the measure and a criterion behaviour at the same time.

  4. Convergent Validity: Examines the extent to which scores on the measure are related to scores on other measures of the same/similar construct.

  5. Discriminant Validity: Opposing constructs should have an opposite relationship, discriminate between construct being measured and unrelated constructs.

Others:

  1. Internal validity:

    • Ability to draw conclusions about the causal relationship from the results of a study.

    • A study has high internal validity when strong inferences can be made that one variable caused changes in the other variable .

    • Analysis of 3 elements:

      1. Temporal Precedence: IV before DV.

      2. Covariation: Both groups sharing condition; DV systematically changes as IV changes.

      3. Eliminate plausible alternatives: The difference is attributable to the IV.

  2. External Validity:

    • Extent to which the results can be generalized to other broader contexts.

    • 2 types:

      1. Population validity: How well the results generalize to the wider population.

      2. Ecological validity: How well the results generalize to real life settings.

Reliability:

  • The consistency or stability of a measure of behaviour.

  • True score: A true value on a real scale.

  • Measurement error: Difference between a true score and a measured score is measurement error.

  • A reliable measure of intelligence will yield similar results each time the individual is measured.

Measuring Reliability:

  • Pearson-Product Moment Correlation Coefficient: This Correlation Coefficient (r) can range from 0.00 to (+/-) 1.00.

  • Correlation of 0.00 tells us that the 2 variables are not related at all; lower reliability. The closer a correlation is to 1.00, the stronger is the relationship; higher reliability.

  • Obtain 2 scores on the measurement, if results are consistent the measure is reliable.

  • A Pearson correlation coefficient that relates the two scores should be a high positive correlation.

  • Test-retest reliability: Assessed by measuring the same individuals at two points in time.

  • Internal Validity: Assess reliability by measuring individuals at only one point in time.

  • Inter-rater Reliability: The extent to which 2 raters agree in their observations.

Measures:

Reactivity of Measures:

Awareness of being measured changes the measured behaviour.

To minimize reactivity: Allow individuals to become used to the presence of observer or the equipment, or use non-reactive measures.

Levels of Measures:

  1. Nominal Scales:

    • No numerical or quantitative properties.

    • Categories or groups simply differ from one another.

    • Also called categorical variables.

  2. Ordinal Scales:

    • Allows to rank order the levels of the variable being studied.

    • The categories can be ordered from first to last.

  3. Interval Scales:

    • Difference between the numbers on the scale, specifically intervals. between the numbers are equal in size.

    • No absolute zero.

  4. Ratio Scales:

    • Has an absolute zero point indicating the absence of the variable being measured.

    • Examples: physical measures, length, weight, or time.