Stats vocab
Review Strategies and Overview
PSY 110 Basic Concepts Overview:
Review all basic terminology from Chapter 1 of the textbook.
Before revealing definitions or examples, recall each definition from memory and generate a original example.
For new terms or concept refreshers, create physical flashcards or utilize a digital flashcard app.
Population and Sample
Population:
Definition: The entire collection of people, animals, or items under consideration or of interest for study. Members of a population share one or more defined attributes or characteristics.
Example 1: All California State University (CSU) undergraduate students (sharing the characteristics of attending a CSU campus and holding undergraduate status).
Example 2: All adults in the United States of America (USA) who identify as Middle Eastern and North African or Southwest Asian and North African (MENA/SWANA).
Example 3: The entire PSY 110 class.
Sample:
Definition: A subgroup drawn from a population. Sampling is conducted because accessing an entire population is frequently impossible, inaccessible, or excessively costly in terms of time, energy, financial capital, and resources.
Example 1: All undergraduate students attending California State University, Fullerton (CSUF), California State University, Chico (CSUC), and California State University, Long Beach (CSULB) serving as a sample of the total CSU undergraduate population.
Example 2: A sample of survey respondents residing in the USA who identify as MENA/SWANA.
Example 3: A pair of students selected from the PSY 110 class.
Sampling Evaluation:
Sampling techniques differ in effectiveness and trade-offs. Assessing sample quality requires evaluating whether the sample is representative of the population, whether the sample size is sufficiently large relative to the population, and whether random sampling methods were implemented.
Parameter and Statistic
Parameter:
Definition: A numerical value or measured characteristic that describes an entire population.
Mnemonic Device: Parameter aligns with Population.
Example 1: If all graduate students at CSULB constitute the population of interest, the average age of all CSULB graduate students (e.g., ) is a parameter.
Example 2: If all adults in Los Angeles (LA) County constitute the population, the median annual income of is a parameter.
Example 3: If all students enrolled at CSULB represent the population, the average height of all CSULB students is a parameter.
Statistic:
Definition: A numerical value or measured characteristic that describes a sample (a subgroup of a population).
Mnemonic Device: Statistic aligns with Sample.
Example 1: The average age of PSY 110 graduate students at CSULB (e.g., ) is a statistic when treating this class as a sample of the broader CSULB graduate student population.
Example 2: If all adults in LA County are the population, the median income calculated from a sample of adults in LA County (e.g., ) is a statistic.
Example 3: If all CSULB students represent the population, the average height calculated from students in a single class is a statistic.
Descriptive and Inferential Statistics
Descriptive Statistics:
Definition: Statistical procedures primarily aimed at summarizing, organizing, and describing data efficiently.
Example 1 (Numerical): Calculating a sample mean, such as determining that the average number of weekly homework assignments completed by graders in a sample is .
Example 2 (Visual): Constructing a frequency graph to display parent education levels.
Inferential Statistics:
Definition: Statistical procedures used to draw conclusions, generalizations, estimations, predictions, or decisions about a population based on sample data, accounting for the inability to measure every member of the population.
Example: Making the claim: "Based on a sample of first graders, evidence suggests that the entire population of first graders, on average, will complete about six homework assignments in one week."
Sampling Error: Inferential tests specifically evaluate and control for sampling error—the natural discrepancy between a sample statistic and its corresponding population parameter—to determine if making the inference from sample to population is justified.
Variables and Data Sets
Variable:
Definition: A specific measurement, characteristic, score, or property that varies or changes across individual people, animals, or objects.
Examples: Car color, age at driver's license acquisition, rat gender, infant birth weight, parenting style, self-efficacy rating, number of class absences.
Constant: A property or measurement that does not vary across individuals or cases.
Data Set:
Definition: A structured collection of multiple variables, typically organized in a tabular format (such as a spreadsheet or statistical software database) where rows represent individual participants/cases and columns represent different variables.
Example 1: An SPSS data file containing rows of participants with columns for name, gender, age, happiness scale score, blood pressure, and other measures.
Example 2: A course gradebook where rows represent individual students and columns represent scores on Quiz 1, Assignment 1, Exam 1, and other assessments.
Latent Constructs and Manifest Variables
Latent Construct:
Definition: An unobservable or indirectly measured psychological attribute or theoretical concept (also referred to as a hypothetical construct). It cannot be directly seen or measured with physical instrumentation and must be inferred from observable responses, self-reports, behaviors, or scale items.
Examples: Intelligence, depression, social anxiety.
Behavioral Inference: Social anxiety is a latent construct because it cannot be physically seen; instead, it is inferred through observable behaviors such as avoiding eye contact, hesitating to join ongoing conversations, or physically stepping back from group activities.
Manifest Variable:
Definition: A variable that can be directly observed and measured.
Examples: Height, eye color, bacterial growth rate, sleep duration.
Discrete and Continuous Variables
Discrete Variable:
Definition: A variable consisting of separate, distinct, countable numerical values with no possible intermediate values existing between two adjacent points (e.g., between and ).
Examples: Number of students in a classroom, number of children in a family, possible outcomes from rolling a standard die, total number of books purchased.
Continuous Variable:
Definition: A variable that can theoretically assume an infinite number of intermediate values between any two points along a continuum and can be measured to an arbitrary degree of precision.
Examples: Weight (measured in kilograms or pounds), height (measured in inches), time duration (measured in minutes).
Predictor, Criterion, and Operational Definitions
Criterion Variable:
Definition: The outcome or response variable evaluated in non-experimental or correlational research designs.
Predictor vs. Criterion Example: In a non-experimental study analyzing the relationship between county of residence and political party affiliation, county serves as the predictor variable and political party affiliation serves as the criterion (outcome) variable.
Operational Definition:
Definition: The exact, concrete procedure or metric used to measure or manipulate a variable within a specific research study. Operational definitions vary across studies and are detailed in the Method section of empirical papers.
Operationalizing Latent Constructs (e.g., Depression): Can be operationally defined as a score on the Beck Depression Inventory, a score on the Minnesota Multiphasic Personality Inventory (MMPI), or a formal diagnostic evaluation by a clinical psychologist.
Operationalizing Manifest Variables (e.g., Height): Can be operationally defined as height measured with a standard measuring tape in inches, or height measured with a yardstick in yards.
Measurement Scales (NOIR and NOS)
System Acronyms:
NOIR: Nominal, Ordinal, Interval, Ratio.
NOS (SPSS Classification): Nominal, Ordinal, Scale (SPSS combines Interval and Ratio into a single category named "Scale").
Nominal Scale:
Properties: Categorical variable consisting of mutually exclusive groups or named categories with no inherent quantitative magnitude, rank, or numerical ordering.
Numerical Coding: Numbers assigned to categories function strictly as qualitative labels without numerical value.
Examples: Hair color (, , , , ), gender, academic major, type of exercise, job industry.
Ordinal Scale:
Properties: Categorical variable with levels ranked or ordered along a continuum indicating varying degrees of a characteristic (least to greatest / smallest to largest). Numbers