Psychology Measurement and Statistics Quiz

Why We Need Statistics

Statistics help us understand the world through both description, summarising data, and inference, drawing conclusions from observation
Statistics are particularly important for psychologists, helping them interpret and assess the meaning of their measurements

Measurement

Psychology often focuses on abstract concepts like intelligence, aggression, or neuroticism which are difficult to measure directly
Operationalization is the process of defining and measuring these abstract concepts through concrete methods, like IQ tests for intelligence.
The validity of an operationalization refers to how well it accurately reflects the underlying abstract concept.

Concepts and Constructs

A concept is a general idea or abstraction, difficult to define precisely (e.g., intelligence)
A construct is a concept with scientific purpose, meaning it can be measured and studied (e.g., IQ)

Terms Review

Variable: A condition, event, characteristic, or attribute that can change in value or category (e.g., age, temperature, intelligence)
Constant: A factor that remains the same in a specific context, not changing or varying (e.g., the number of hours in a day)

Experimental Terms

Independent Variable (IV): Variable directly manipulated by the researcher, considered the "cause" in an experiment; "influence" in non-experiments (e.g., type of therapy in a depression study)
Dependent Variable (DV): Variable being measured or observed, considered the outcome, "effect" or "result" (e.g., depression symptoms severity in a therapy study)

Non-Experimental Terms

Predictor Variable: Variable believed to influence another variable in a non-experimental setting (e.g., personality traits predicting job performance)
Criterion Variable: The variable being predicted or influenced by another variable, often the outcome variable (e.g., job performance in a study about personality traits)

Other Terms

Binary (Dichotomous) Variable: Variable with only two possible values (e.g., male/female, yes/no)
Dummy Variable: Variable created by recoding a categorical variable with multiple categories into a series of binary variables (e.g., recoding "race" into "white/non-white")
Endogenous Variable: Variable inherent to the system being studied, determined from within the system, often caused by other variables within the system (e.g., in a model of economic growth, the level of investment could be an endogenous variable influenced by variables like interest rates)
Confounding Variable: Variable that obscures the relationship between other variables, making it difficult to determine true cause and effect (e.g., the skill level of the experimenter might influence results in a study, confounding the effects of the treatment)
Control Variable: Variable deliberately kept constant to eliminate its potential influence on the relationship between other variables (e.g., controlling for age in a study on the effects of medication)
Exogenous Variable: Variable external to the system being studied, not affected by the variables within the system (e.g., in a study of economic growth, a shock to the oil price would be an exogenous variable)
Latent Variable: Underlying variable that cannot be directly observed but is inferred from other observable variables (e.g., anxiety cannot be measured directly but can be inferred from behavioral observations)
Manifest Variable: Observable variable that is assumed to reflect the presence of a latent variable (e.g., sweating, rapid heart rate, and trembling could be manifest variables that indicate anxiety)
Mediating Variable: Variable that explains the relationship between two other variables by mediating the causal pathway (e.g., stress can mediate the relationship between demanding jobs and depression, explaining how demanding jobs lead to stress, which in turn leads to depression)
Moderating Variable: Variable that influences the relationship between two other variables, creating an interaction effect. The relationship between the first two variables changes depending on the level of the moderating variable (e.g., the relationship between exercise and health might be moderated by age, meaning exercise has a stronger effect on health for younger people)
Polychotomous Variable: Variable with more than two possible values, encompassing all variables except binary variables (e.g., "marital status" with categories "single," "married," "divorced," "widowed")

Frequency Distribution

A simple frequency distribution uses individual scores to show how often each score occurs in a data set.
A grouped frequency distribution uses class intervals, or ranges of scores, to show how many scores fall within each interval.
Percentiles: Specific scores or points within a distribution that divide the data into hundredths, indicating what percentage of scores fall below that score.
Percentile Ranks: Indicate the percentage of scores that fall below a particular score.

Measures of Central Tendency

A measure of central tendency summarizes a distribution of scores by indicating the average or middle value.
Mean: Arithmetic average, calculated by summing all scores and dividing by the number of scores.
Median: The middle score when the data is ranked in order.
Mode: The most frequently occurring score in a distribution.

Measures of Variability

Measures of variability describe how spread out the scores are in a distribution.
Range: The difference between the highest and lowest scores in a distribution.
Interquartile Range: The difference between the third quartile (Q3) and the first quartile (Q1), capturing the middle 50% of the data.
Semi-Interquartile Range: Half of the interquartile range, providing information about the spread of scores around the median.
Average Deviation: Rarely used, due to its inadequacy in further calculations.
Standard Deviation: The square root of the variance, indicating the typical distance between each score and the mean. It is a common measure of variability.

The Normal Curve

The normal curve is a bell-shaped, symmetrical distribution with most scores clustered around the center.
It is also known as the "Laplace-Gaussian curve" and is a theoretical ideal.

Measures of Shape

Skewness: Describes the asymmetry of a distribution. A distribution is positively skewed when there are fewer high scores and negatively skewed when there are fewer low scores.
Kurtosis: Describes the steepness of a distribution at its center.