final - research methods

Positivism

The philosophical pillar of quantitative research assuming social reality is objective and can be measured to discover universal causal laws

Deductive Reasoning

Research moving from the general to the specific; it begins with a theoretical pattern and tests it through observations

Variable

A logical grouping of attributes (e.g., "Gender")

Attribute

A characteristic or quality describing an object or person (e.g., "Female")

Independent Variable (IV)

The cause in a relationship; the variable expected to determine an attribute in another

Dependent Variable (DV)

The effect; the variable that depends on or is influenced by the independent variable

Control Variable

Used to generate an unbiased image of the world by maintaining controlled conditions

Spurious

A relationship where an unseen third variable is the actual cause of both the independent and dependent variables

Nomothetic Research

Aims to explain a class of situations efficiently using a few factors, settling for a partial rather than exhaustive explanation

Idiographic Research

Aims to understand a single situation exhaustively, focusing on its unique causes

Hypothesis

An informed, measurable prediction about a relationship between variables based on existing literature

Null Hypothesis

A statement assuming no relationship exists between variables

Falsifiable

A requirement that a hypothesis must be capable of being proven wrong through empirical testing

Unit of Analysis

The "who" or "what" being studied (e.g., individuals, groups, or organizations)

Ecological Fallacy

An error in explanation where researchers make assertions about individuals based on group-level data

Reductionism

An error where complex social phenomena are reduced to a single, simple cause

Directional Hypothesis

Specifies the nature of the relationship (e.g., "Group A will be higher than Group B")

Non-directional Hypothesis

Predicts a relationship exists but does not specify the direction

Confounding Variable

An extraneous factor that can hide or distort the relationship between the independent and dependent variables

Conceptual Definition

An abstract, theoretical idea a researcher seeks to study

Operational Definition

The specific way a concept is measured, turning it into a numerical variable

Empirical Indicators

Specific, observable signs used to measure a concept

Categorical (Discrete) Variable

Assigns observations into particular categories (Nominal/Ordinal)

Continuous Variable

A numerical measure that can take any value within a range (Interval/Ratio)

Nominal Variable

Categories with no ranked order (e.g., gender, marital status)

Ordinal Variable

Categories that are ranked, but the distance between them is unknown (scales)

Interval Variable

Numerical distance between points is equal, but there is no true zero (e.g., IQ scores, years)

Ratio Variable

Precise measurement with a true zero (e.g., income, age in years)

Index

A composite measure that summarizes several specific observations

Scale

A composite measure that captures the intensity or degree of a concept

Face Validity

Whether a measure appears, on the surface, to measure what it claims to

Content Validity

How much a measure covers the full range of meanings within a concept

Criterion Validity

Checking a measure against some external criterion (predictive or concurrent)

Probability Sampling

Every member of the population has an equal and known chance of selection

Simple Random

Every unit has an equal chance via random selection (e.g., computer generator)

Systematic

Selection based on a mathematical algorithm (e.g., every 10th person)

Stratified

Dividing the population into subgroups (strata) and sampling proportionally

Cluster

Sampling naturally occurring groups (e.g., schools) rather than individuals

Non-Probability Sampling

Non-random selection with no known representativeness

Convenience

Choosing participants who are easiest to reach

Purposive

Selecting participants based on specific characteristics

Quota

Selecting a pre-set number of people from specific categories

Snowball

Participants recruit other participants from their social networks

Sampling Frame

The actual list of units from which the sample is drawn

Sampling Element

The individual unit of the population that is selected

Representativeness

The degree to which a sample accurately reflects the population

Generalizability

The extent to which findings can be applied to the broader population (external validity)

Open-ended

Respondents answer in their own words

Close-ended

Respondents choose from provided, mutually exclusive, and exhaustive categories

Double-Barreled

Asking two questions but allowing only one answer

Leading

Questions that suggest a "correct" answer

Loaded

Use of emotionally charged language to bias responses

Vignette

Using a hypothetical story to elicit a response

Contingency/Filter

Questions used to determine if a respondent should answer a subsequent set of questions

Social Desirability Bias

Answering in a way that makes the respondent look good rather than being truthful

Wording Effects

How the specific phrasing of a question influences the answer

Context Effects

How the order or surrounding questions influence a response

True Experiment

Requires an experimental group, a control group, and random assignment

Random Assignment

Dividing subjects into groups using a random process to ensure they are comparable

Hawthorne Effect

Participants change behavior because they know they are being studied

Pretest/Post-test

Measurements taken before and after an intervention

Selection Bias

Error resulting from differences between those who take part in a study and those who do not

Mortality (Attrition)

Participants dropping out of a study over time

Maturation

Changes in subjects over time that happen naturally and are not caused by the intervention

Univariate Statistics

Analysis of a single variable at a time

Multivariate Statistics

Analysis of more than two variables simultaneously

Measures of Central Tendency

• Mean: The mathematical average.

• Median: The middle value in a distribution.

• Mode: The most frequent value

Standard Deviation

A measure of dispersion; high values mean data is spread out, low values mean it is clustered near the mean

Skewness

• Positive (Right) Skew: High-value outliers pull the mean to the right [history, 643].

• Negative (Left) Skew: Low-value outliers pull the mean to the left

Errors in Significance

• Type I Error: Rejecting a null hypothesis that is actually true ("false positive").

• Type II Error: Failing to reject a null hypothesis that is actually false ("false negative")

Stakeholders

People in an organization invested in the program being evaluated

Prospective Cohort Study

Following a group into the future to observe outcomes

Retrospective Cohort Study

Looking back at past data for a specific group to determine outcomes

Formative Evaluation

Focused on improving program implementation and function while it is running

Summative Evaluation

Conducted post-project to determine if goals and outcomes were achieved

Needs Assessment

Identifying unmet needs and gaps in service to prioritize resources

Evaluability Assessment

Determining if a program is capable of being evaluated at all

Mixed Methods

Systematic integration of quantitative and qualitative data in a single project

Investigator

Using multiple researchers to observe the same phenomenon

Theoretical

Using different theoretical perspectives to interpret one dataset

Methodological

Using different data collection methods (e.g., survey and interview)

Analytical

Using different ways to analyze the same data

Concurrent

Carrying out components at approximately the same time

Sequential

Carrying out one component after another (e.g., QUAL → QUANT)

Terra Nullius / Datum Nullius

Historically used concepts of "no man's land" applied to data, suggesting it belongs to no one and can be taken

indigenous Data Sovereignty

The right of Indigenous peoples to govern the collection, ownership, and use of data about them

OCAP Principles:

• Ownership: Communities own their information.

• Control: Communities control all aspects of research.

• Access: Communities have access to data about themselves.

• Possession: Communities physically possess the data

Statistics as Culturally Embedded

Recognition that numerical data is not neutral but reflects the cultural values and biases of those who collect it

5D Data

A framework for Indigenous data including: Disaggregated, Diverse, Decolonized, Distinction-based, and Data-sovereign

Data-rich vs. Data-poor Actors

Disparities in who has access to high-quality data and information resources

Culture as Cure

Using traditional cultural knowledge and practices as the foundation for community wellness and reintegration

Whakapapa

A Māori term referring to genealogy or the layers of relationships and history that connect people to place and each other