Quantitative Research Glossary

Key quantitative and statistical terms, theories, symbols, and tools used in this dissertation.

Theories and Models

Core theoretical models and frameworks that inform technology adoption and behavioral intentions.

Unified Theory of Acceptance and Use of Technology (UTAUT)

A technology acceptance model developed by Venkatesh et al. (2003) that integrates eight prior models to explain technology usage through four core constructs: performance expectancy, effort expectancy, social influence, and facilitating conditions.

Unified Theory of Acceptance and Use of Technology 2 (UTAUT2)

An extension of UTAUT that adds hedonic motivation, price value, and habit as additional predictors of behavioral intention and use, tailored to individual and consumer contexts. UTAUT2 is the primary theoretical framework for this dissertation’s examination of cyber incident response plan (CIRP) adoption.

Theory of Reasoned Action (TRA)

A social psychology model positing that behavior is driven by behavioral intentions, which are influenced by attitudes toward the behavior and subjective norms.

Technology Acceptance Model (TAM)

A technology adoption model in which perceived usefulness and perceived ease of use determine a person’s attitude toward using a system, which then influences behavioral intention and use.

Theory of Planned Behavior (TPB)

An extension of TRA that adds perceived behavioral control as a determinant of behavioral intention and behavior, acknowledging constraints and facilitators beyond attitude and norms.

Combined TAM–TPB (C-TAM-TPB)

An integrated model that merges constructs from TAM and TPB to predict technology adoption by including usefulness, ease of use, attitude, norms, and perceived control.

Model of PC Utilization (MPCU)

A model that explains PC usage behavior through factors such as job fit, complexity, long-term consequences, and social factors, and is one of the eight models synthesized into UTAUT.

Innovation Diffusion Theory (IDT)

A theory describing how innovations spread through populations over time, emphasizing characteristics such as relative advantage, compatibility, complexity, trialability, and observability.

Social Cognitive Theory (SCT)

A theory that explains behavior as a dynamic interaction of personal factors, behavior, and environment, with emphasis on self-efficacy and observational learning.

Cyber Incident Response Plan (CIRP)

A formal, strategic document that defines responsibilities, procedures, and communication protocols for preparing for, managing, and recovering from cybersecurity incidents within an organization.

Greek Symbols and Notation in PLS-SEM

Detailed explanations of Greek symbols as they are used within Partial Least Squares Structural Equation Modeling (PLS-SEM).

λ (Lambda) — Indicator Loading / Outer Loading

In PLS-SEM, λ represents the outer loading, the strength of the relationship between an indicator (e.g., PE1, FC3) and its latent construct. Loadings ≥ 0.70 indicate strong convergent validity. Low λ values suggest the indicator should be reviewed or removed.

β (Beta) — Path Coefficient

β is the standardized path coefficient in the structural model, indicating the strength and direction of the relationship between latent variables (e.g., PE → BI). Values closer to ±1 indicate stronger relationships.

α (Alpha) — Cronbach’s Alpha / Significance Level

Most commonly used as Cronbach’s alpha, a measure of internal consistency reliability of indicators within a construct. Values ≥ 0.70 indicate acceptable reliability.
α is also used as the significance level in hypothesis testing (usually 0.05).

ρ_A (Rho_A) — Construct Reliability

A reliability measure preferred in PLS-SEM because it uses the actual outer loadings. It lies between Cronbach’s alpha and Composite Reliability. Values ≥ 0.70 indicate adequate reliability.

ρ_C (Composite Reliability, CR)

Composite Reliability is another measure of internal consistency in PLS-SEM. Values above 0.70 indicate good reliability, and it is less biased than α because it uses item loadings rather than assuming equal weight.

φ (Phi) — Correlation Between Latent Variables

Phi represents the correlation between two latent constructs in the model before structural paths are applied. It appears in early model diagnostics and sometimes in discriminant validity evaluation.

δ (Delta) — Measurement Error (Reflective Indicators)

Represents the error variance in reflective indicators. Since PLS-SEM focuses on maximizing explained variance, δ = 1 − λ². Higher δ indicates more measurement error.

ε (Epsilon) — Structural Error Term / Residuals

Represents the unexplained variance in the dependent latent variable in the structural model. In BI = f(PE, FC, …), ε cap

---

---

Statistical and PLS-SEM Terminology

Collinearity

Collinearity occurs when two predictor variables are highly correlated with each other. In PLS-SEM, high collinearity can inflate standard errors and distort path coefficients. It is assessed using variance inflation factor (VIF); values above 3.3 or 5.0 may indicate problematic collinearity.

Multicollinearity

Multicollinearity refers to high correlations among three or more predictor variables. Severe multicollinearity weakens the statistical ability to isolate individual predictor effects. PLS-SEM uses VIF diagnostics at both the indicator and construct levels.

Covariance

Covariance measures how two variables change together. A positive covariance indicates that both variables move in the same direction, while a negative covariance indicates opposite movement. Covariance is a foundational concept in SEM, measurement models, and reliability analysis.

Correlation

A standardized measure of the relationship between two variables, ranging from -1 to +1. Correlation does not imply causation but indicates the strength and direction of association.

Standard Deviation (SD)

The standard deviation quantifies how much individual scores deviate from the mean. A higher SD indicates greater variability in responses. SD is commonly reported in descriptive statistics tables in PLS-SEM and SPSS outputs.

Variance

The average squared deviation from the mean. Variance is the foundation for SD, reliability coefficients, and measurement model evaluation.

Composite Reliability (CR)

A reliability measure used in PLS-SEM to assess internal consistency of latent constructs. Acceptable values range from 0.70 to 0.95. CR is preferred over Cronbach’s alpha when evaluating reflective measurement models.

Cronbach’s Alpha

A classic reliability measure assessing the internal consistency of a scale. Although widely used, PLS-SEM researchers often rely more on composite reliability and rho_A for accuracy.

AVE (Average Variance Extracted)

AVE evaluates convergent validity of a reflective construct. It represents how much of the variance in indicators is explained by the latent variable. A value ≥ 0.50 is required.

Outer Loadings

Indicator-to-construct correlations in PLS-SEM. Loadings above 0.70 are desirable, though values as low as 0.40 may be retained depending on AVE and theory.

Path Coefficient

Represents the strength and direction of the relationship between constructs in the structural model. Path coefficients are interpreted similarly to standardized betas in regression.

Bootstrapping

A resampling method used in PLS-SEM to estimate significance levels for path coefficients, weights, and loadings. Confidence intervals and p-values come from bootstrap procedures.

t-value

A statistic generated from bootstrapping that determines whether a path or loading is statistically significant. Higher t-values indicate stronger evidence against the null.

p-value

Indicates the probability that the observed effect occurred by chance. In PLS-SEM, p < .05 is generally considered statistically significant.

R² (Coefficient of Determination)

R² indicates how much variance in an endogenous construct is explained by its predictors. Ranges: 0.25 = weak, 0.50 = moderate, 0.75 = substantial (Hair et al.).

Effect Size (f²)

Measures the impact of one predictor on an endogenous construct by evaluating the change in R² when the predictor is removed. Thresholds: 0.02 (small), 0.15 (medium), 0.35 (large).

Predictive Relevance (Q²)

Assessed using blindfolding; Q² > 0 indicates that the model has predictive relevance for the construct.

Discriminant Validity

Assesses whether constructs are distinct from one another. Common criteria include the Fornell–Larcker criterion, cross-loadings, and HTMT (heterotrait–monotrait ratio).

HTMT (Heterotrait–Monotrait Ratio)

A modern discriminant validity metric. HTMT values should be below 0.85 (strict) or 0.90 (liberal).

Latent Variable

An unobserved conceptual variable measured indirectly through indicators (e.g., performance expectancy).

Indicator

A survey item or observed variable used to measure a latent variable.

Structural Model

The part of the PLS-SEM model that specifies relationships among latent variables, representing hypothesized causal paths.

Measurement Model

Specifies how latent variables are measured through their indicators. Includes reflective and formative models.

---

---

Inferential Statistics Terminology

Null Hypothesis (H0)

A statement asserting that there is no relationship or effect between variables. Hypothesis testing begins with the assumption that the null hypothesis is true.

Alternative Hypothesis (H1)

A statement proposing that a relationship or effect does exist. Research hypotheses typically align with the alternative hypothesis.

Significance Level (α)

The threshold for rejecting the null hypothesis, commonly set at 0.05. Values below α indicate statistically significant results.

Confidence Interval (CI)

A range of values within which the true population parameter is expected to fall with a specified level of confidence (typically 95%).

Standard Error (SE)

The standard deviation of a sampling distribution. SE decreases as sample size increases and is fundamental for constructing confidence intervals.

t-Test

A statistical test used to compare means or determine whether an observed effect differs significantly from zero. In PLS-SEM, t-tests are produced through bootstrapping.

Degrees of Freedom (df)

A value representing the number of independent pieces of information available to estimate a parameter. Commonly used in t-tests, chi-square tests, and ANOVA.

Chi-Square (χ²)

A statistic used to test relationships between categorical variables or to assess model fit in covariance-based SEM. Not used as a fit index in PLS-SEM.

ANOVA (Analysis of Variance)

A statistical procedure testing whether mean differences exist across groups. Based on the F-statistic, comparing between-group and within-group variation.

F-Statistic

A ratio of explained variance to unexplained variance in ANOVA and regression models. A higher F-value indicates a more statistically significant model.

Regression Coefficient (β)

Represents the strength and direction of the relationship between a predictor and an outcome. In PLS-SEM, equivalent to path coefficients.

Residual

The difference between observed values and predicted values. Residual patterns are used to assess assumptions such as linearity and homoscedasticity.

Homoscedasticity

The assumption that residuals have equal variance across all levels of the predicted values. Violations may bias statistical conclusions.

Heteroscedasticity

Unequal variance in residuals across predicted values. Can affect accuracy of coefficients and significance testing.

Normality

The assumption that data follow a normal distribution. While PLS-SEM does not require multivariate normality, many classical inferential tests assume it.

Linearity

The assumption that relationships between variables are linear. Checked using scatterplots or residual diagrams.

Outlier

A data point significantly distant from others in the distribution. Outliers may distort correlations and regression results and are commonly evaluated before analysis.

Effect Size

A measure of the magnitude of an observed effect, independent of sample size. Common effect sizes include Cohen’s d, f², and R².

Power (Statistical Power)

The probability of correctly rejecting a false null hypothesis. Higher power reduces the risk of Type II errors. Power increases with sample size.

Type I Error

Incorrectly rejecting the null hypothesis (a false positive). Controlled by the significance level α.

Type II Error

Failing to reject a false null hypothesis (a false negative). More likely with small sample sizes.

Parametric Test

A statistical test that assumes certain conditions about the population distribution (e.g., t-test, ANOVA).

Nonparametric Test

A distribution-free test used when parametric assumptions are not met (e.g., Mann–Whitney U test).

---

---

SPSS Terminology

Data View

The spreadsheet-like interface in SPSS where each row represents a case (participant) and each column represents a variable. This is where raw data are displayed.

Variable View

The interface where variables are defined. Each row represents a variable and includes settings such as name, label, type, values, missing data codes, and measurement level.

Measurement Levels

SPSS uses four levels: nominal (categories), ordinal (ranked categories), scale (interval/ratio), and string (text). Scale variables are used for correlations, regression, and most inferential tests.

Descriptive Statistics

A table summarizing the mean, standard deviation, minimum, maximum, and sample size (N). Frequently used as the first output block in quantitative studies.

Frequencies

Tables showing the count and percentage of each category for nominal or ordinal variables such as gender, age groups, or education level.

Explore Function

An SPSS tool used to generate descriptive statistics, boxplots, and normality tests for assessing distribution characteristics and spotting outliers.

Pearson Correlation Table

A matrix showing correlations (r), significance levels (p-values), and sample sizes (N). Appears commonly in preliminary analysis sections to explore linear relationships among variables.

Sig. (2-tailed)

The p-value associated with a correlation or test statistic. If < 0.05, the relationship is typically considered statistically significant.

Reliability Analysis

SPSS procedure used to compute Cronbach’s alpha, item-total correlations, and “alpha if item deleted.” Commonly used for assessing reliability of survey scales.

Cronbach’s Alpha Table

Output showing the internal consistency of a set of items. Values above 0.70 indicate acceptable reliability for research use.

Item-Total Statistics

A table showing how each item correlates with the total scale and how alpha changes if an item is removed. Useful for identifying weak items.

Missing Values

SPSS supports three types of missingness: user-missing (coded by researcher), system-missing (empty), and excluded cases listwise or pairwise in analysis.

Listwise Deletion

Excludes a participant entirely from an analysis if they have missing data for any variable used in that test. Ensures consistency but lowers sample size.

Pairwise Deletion

Uses all available data and excludes participants only when they are missing values relevant to a specific calculation. Maintains larger N but can complicate interpretation.

Standardized Residuals

Residuals scaled into standard deviation units. Useful for identifying outliers or assessing model fit.

Scatterplot

Visual tool used for inspecting linearity and spotting potential outliers prior to correlation or regression analysis.

Boxplot

A visual representation of distributions used to identify median, quartiles, and possible outliers.

Histogram

A bar graph used to visually assess distribution shape, skewness, and normality tendencies.

Normal Q-Q Plot

A diagnostic plot comparing observed data to a theoretical normal distribution. Points close to the diagonal line suggest approximate normality.

Shapiro–Wilk Test

A statistical test for normality. A p-value > 0.05 suggests data are not significantly different from a normal distribution.

Levene’s Test

A test for homogeneity of variance (equal variances across groups). Frequently used in ANOVA and t-test procedures.

Regression Output (Coefficients Table)

Contains beta coefficients, t-values, p-values, and collinearity statistics. Central to interpreting predictive relationships in SPSS regression.