Comprehensive Study Guide: Introduction to Psychological Statistics

Core Concepts of Statistics

• Definition of Statistics: Statistics is defined as a set of mathematical procedures used for organizing, summarizing, and interpreting information.

• Primary Purposes of Statistics:

  • Organizing and Summarizing Data: Simplifies large sets of measurements into manageable and understandable formats.

  • Answering Research Questions: Provides analytical tools to evaluate evidence and draw conclusions regarding the questions that initiated the study.

Populations, Samples, Parameters, and Statistics

• Population:

  • The complete set of all individuals, items, or elements of interest in a specific research study.

  • Populations can vary in size from small, highly specific groups to vast, generalized groups.

• Sample:

  • A subset of individuals selected from a population.

  • Intended to accurately represent the population in a research study.

• Parameter:

  • A numerical value that describes an entire population.

  • Derived directly from measurements of all individuals within the population.

• Statistic:

  • A numerical value that describes a sample.

  • Derived from measurements of the individuals within a given sample.

Branches of Statistical Methods

• Descriptive Statistics:

  • Statistical procedures used to summarize, organize, and simplify data.

  • Methods: Tabulating data, creating graphical representations (e.g., bar graphs, pie charts, line plots), and calculating summary measures such as averages or percentages.

• Inferential Statistics:

  • Methods that utilize sample data to make general statements, generalizations, or predictions about an entire population.

  • Accounts for sampling error to determine how reliably sample results reflect population parameters.

Psychological Constructs and Operational Definitions

• Constructs (Hypothetical Constructs):

  • Internal attributes, traits, or characteristics that cannot be directly observed or measured.

  • Useful for describing, explaining, and predicting human behavior.

  • Examples: Intelligence, motivation, anxiety, and fear.

• Operational Definitions:

  • Identifies a precise measurement procedure (a set of operations) for observing and measuring an external, concrete behavior.

  • Uses the resulting numerical or categorical measurement as both a definition and a metric of the underlying hypothetical construct.

• Examples of Operational Definitions:

  • Construct: Intelligence

    • Operational Definition: Score on the Verbal SAT test (a standardized test measure) $\rightarrow$ Value: $550$

    • Operational Definition: Speed of repairing an engine (a behavioral performance measure) $\rightarrow$ Value: $2\text{ hrs}$

    • Operational Definition: Number of hairs on the left thumb (an invalid/flawed operational definition) $\rightarrow$ Value: $7$

  • Construct: Age

    • Operational Definition: Response to a survey questionnaire item (a self-report measure) $\rightarrow$ Value: $21$

Classification of Variables

• Variable:

  • Any characteristic, property, or attribute of an event, object, or person that can assume different values across individuals or across different points in time.

• Data and Raw Scores:

  • Data: The complete set of measurements or observations collected for a variable across individuals.

  • Raw Score / Score: The individual value or measurement obtained for a single subject (denoted as $X$ or $Y$).

• Qualitative (Categorical) Variables:

  • Variables that differ in quality, type, or kind rather than magnitude.

  • Consist of discrete non-numerical categories (e.g., eye color, brand of smartphone).

• Quantitative (Numerical) Variables:

  • Variables that differ in amount, degree, or magnitude, represented by numerical values.

• Discrete Variables:

  • Consist of separate, indivisible categories.

  • No intermediate numerical values can exist between two neighboring categories.

  • Examples: Biological sex (Male/Female), Boolean outcomes (True/False), count of people in a room (finite numbers).

• Continuous Variables:

  • Contain an infinite number of possible values that fall between any two observed values.

  • Infinitely divisible into fractional or decimal components.

  • Examples: Height, weight, distance, time.

Scales (Levels) of Measurement

• Nominal Scale:

  • Consists of a set of named, mutually exclusive categories.

  • Used purely to label, classify, or categorize observations without conveying numerical magnitude or order.

  • Examples: Eye color (Blue, Brown, Green), Smartphone brands (iPhone, Samsung, Moto), Transportation modes (Bus, Train, Car).

  • Statistical Analysis: Frequency distributions, mode, non-parametric statistical tests.

• Ordinal Scale:

  • Consists of categories arranged in a natural ordered sequence or ranking.

  • Ranks observations in terms of relative size, magnitude, or standing, but intervals between ranks are not guaranteed to be equal.

  • Examples: Letter grades (A, B, C), Education levels (Bachelor's, Master's, PhD), Job seniority levels (Junior, Mid, Senior).

  • Statistical Analysis: Frequency distributions, mode, median, range, non-parametric statistical tests.

• Interval Scale:

  • Consists of ordered categories that form equal intervals of magnitude across the scale.

  • The distance between adjacent scale points is constant and known.

  • Has an arbitrary zero point: A score of zero does not represent a complete absence of the variable being measured.

  • Examples: Temperature in Fahrenheit/Celsius ($70^\text{o}$, $80^\text{o}$, $90^\text{o}$), IQ standardized scores ($40$, $100$, $160$), grouped income ranges.

  • Statistical Analysis: Frequency distributions, mode, median, mean, range, standard deviation, variance, parametric statistical tests (e.g., t-test, linear regression).

• Ratio Scale:

  • An interval scale featuring an absolute (true) zero point.

  • A measurement value of zero indicates a complete absence of the measured attribute.

  • Ratios of scale numbers reflect true proportional ratios of magnitude (e.g., $10\text{ kg}$ is twice as heavy as $5\text{ kg}$).

  • Examples: Weight in kilograms ($50$, $70$, $90$), Staff count ($10$, $30$, $50$), Annual income in USD ($\$20\text{k}$, $\$40\text{k}$, $\$60\text{k}$).

  • Statistical Analysis: Frequency distributions, mode, median, mean, range, standard deviation, variance, coefficient of variation, parametric statistical tests (e.g., ANOVA, linear regression).

Statistical Notation and Summation Rules

• Summation Notation ($\Sigma$):

  • The uppercase Greek letter sigma ($\Sigma$) represents mathematical summation.

  • The expression $\Sigma X$ translates as "the sum of all scores for variable $X$".

  • The letter $N$ represents the total count of scores/observations in a population, while $n$ represents the count in a sample.

• Basic Summation Example:

  • Given dataset for variable $X$: $10, 6, 7, 4$

  • Sample size: $N = 4$

  • Calculated sum: $\Sigma X = 10 + 6 + 7 + 4 = 27$

• Summation Problem 1: Squaring Scores vs. Summing Squares

• Computations:

  • Sum of Scores ($\Sigma X$):         $\Sigma X = 3 + 1 + 7 + 4 = 15$

  • Sum of Squared Scores ($\Sigma X^2$):         $\Sigma X^2 = 3^2 + 1^2 + 7^2 + 4^2 = 9 + 1 + 49 + 16 = 75$

  • Square of the Sum ($(\Sigma X)^2$):         $(\Sigma X)^2 = (15)^2 = 225$

• Summation Problem 2: Operations Inside Parentheses

• Computations:

  • Sum of Deviations ($\Sigma (X - 1)$):         $\Sigma (X - 1) = (3 - 1) + (1 - 1) + (7 - 1) + (4 - 1) = 2 + 0 + 6 + 3 = 11$

  • Sum of Squared Deviations ($\Sigma (X - 1)^2$):         $\Sigma (X - 1)^2 = 2^2 + 0^2 + 6^2 + 3^2 = 4 + 0 + 36 + 9 = 49$

• Summation Problem 3: Products of Two Variables

• Computations:

  • Sum of $X$ ($\Sigma X$):         $\Sigma X = 3 + 1 + 7 + 4 = 15$

  • Sum of $Y$ ($\Sigma Y$):         $\Sigma Y = 5 + 3 + 4 + 2 = 14$

  • Sum of Products ($\Sigma XY$):         $\Sigma XY = (3 \times 5) + (1 \times 3) + (7 \times 4) + (4 \times 2) = 15 + 3 + 28 + 8 = 54$