Basics of Measurement and Statistics in Psychological Research

Basics of Measurement and Statistics in Psychological Research

Structure of Statistics

Statistics is a foundational tool in psychological research, enabling researchers to summarize data, test hypotheses, and draw inferences about populations. Understanding the structure and types of statistics is essential for designing and interpreting research studies.

• Types of Statistics:

  • Null Hypothesis Significance Testing (NHST): A method that evaluates whether observed data are likely under a null hypothesis, typically using p-value thresholds to determine statistical significance.

  • Bayesian Statistics: An alternative approach that incorporates prior knowledge or beliefs, updating them with observed data to form posterior probabilities.

• Descriptive Statistics: Methods that summarize and describe the main features of a dataset. Examples include measures of central tendency (mean, median, mode) and variability (range, standard deviation).

• Inferential Statistics: Techniques that allow researchers to make conclusions or inferences about a population based on sample data. This includes hypothesis testing and estimation.

Example: A table summarizing home sales by type (single family, condominium, townhouse) provides descriptive statistics, while comparing average prices between types uses inferential statistics.

Research Design

Research design refers to the structured plan for investigating a research question. It determines how variables are manipulated and measured, and how participants are assigned to conditions.

• Independent Variable (IV): The variable that is manipulated or categorized to observe its effect on another variable. Also called the predictor or X variable.

• Dependent Variable (DV): The outcome variable that is measured to assess the effect of the independent variable. Also called the criterion or Y variable.

• Operational Definitions: Precise descriptions of how variables are measured or manipulated in a study. For example, reaction time might be defined as the number of milliseconds from the start signal to the finish line crossing.

• Population vs. Sample:

  • Population: The entire set of individuals or items of interest in a study.

  • Sample: A subset of the population selected to represent the whole.

  • Sampling Error: The discrepancy between a sample statistic and the corresponding population parameter, due to random variation.

Example: To study the effect of noise on studying, the IV could be room noise level (quiet vs. loud), and the DV could be test scores.

Numbers and Measurement

Measurement in psychology involves assigning numbers to represent attributes or properties of objects, individuals, or events according to specific rules. The type of numbers and scales used determines the appropriate statistical analyses.

• Numbers as Codes: Numbers can serve as symbolic classifications (e.g., 1 = Canada, 2 = United States, 3 = China) without quantitative meaning.

• Continuous vs. Discrete Numbers:

  • Discrete: Numbers with a finite set of possible values (e.g., number of students in a class).

  • Continuous: Numbers that can take any value within a range (e.g., height, reaction time).

• Measurement Error: The difference between the observed value and the true value. It can be systematic (consistent bias) or random (unpredictable variation).

  • Equation:

  $X_{observed} = X_{true} + X_{error}$

  • Systematic Error: Consistent, repeatable error associated with faulty equipment or bias.

  • Random Error: Error that varies unpredictably from one measurement to another.

Example: If a scale is always 2 kg too heavy, this is a systematic error. If the scale fluctuates randomly, this is random error.

Scales of Measurement

Scales of measurement define the relationship between the numbers assigned to objects and the properties being measured. The scale type determines which statistical analyses are appropriate.

• Nominal Scale: Categorical data where numbers serve as labels without quantitative value. Categories are distinct but unordered.

  • Example: Types of pets (1 = dog, 2 = cat, 3 = bird).

• Ordinal Scale: Data with a meaningful order but unknown intervals between values.

  • Example: Olympic medals (gold, silver, bronze).

• Interval Scale: Ordered data with equal intervals between values, but no true zero point.

  • Example: Temperature in Celsius or Fahrenheit.

• Ratio Scale: Ordered data with equal intervals and a true zero point, allowing for statements about proportions.

  • Example: Height, weight, reaction time.

Scale	Order	Equal Intervals	True Zero	Example
Nominal	No	No	No	Gender, Country
Ordinal	Yes	No	No	Rankings, Medals
Interval	Yes	Yes	No	Temperature (°C)
Ratio	Yes	Yes	Yes	Height, Weight

Additional info: The choice of scale affects which statistical tests are valid. For example, means and standard deviations are meaningful for interval and ratio data, but not for nominal data.

Descriptive and Inferential Statistics: Applications

Both descriptive and inferential statistics are used to analyze psychological data, but they serve different purposes.

• Descriptive Statistics: Used to summarize data from a sample using measures such as mean, median, mode, and standard deviation.

• Inferential Statistics: Used to make generalizations from a sample to a population, often involving hypothesis testing (e.g., t-tests, ANOVA, correlation).

Example: Calculating the average test score in a class (descriptive), then using a t-test to compare scores between two classes (inferential).

Summary Table: Types of Statistics

Type	Main Purpose	Example
Descriptive	Summarize and describe data	Mean, median, mode, standard deviation
Inferential	Draw conclusions about populations	t-test, ANOVA, correlation, regression