This paper demonstrates how to conduct a hierarchical multiple regression analysis using SPSS to identify predictors of resilience. It outlines the variables entered into the analysis — including demographic factors, protective factors, childhood socioeconomic status, and childhood maltreatment — and explains how each step of the hierarchical model contributes to explaining variance in resilience composite scores. The paper walks through the SPSS Variable View and Data View setup, distinguishes between categorical and scale variables, and discusses how the output is interpreted to determine which variables uniquely predict resilience beyond the contributions of previously entered predictors.
The paper demonstrates hierarchical multiple regression, a method that builds a statistical model in sequential blocks to assess the incremental predictive contribution of each group of variables. By entering demographics first, then protective factors, then socioeconomic and maltreatment variables, the analyst can isolate which predictors explain unique variance in resilience above and beyond prior blocks — a sophisticated approach to causal modeling in social science research.
The paper opens by listing the full variable set, then explains what each variable represents and how it was measured. It introduces the dependent variable and rationale for the regression, then walks through the four hierarchical steps. It closes by referencing SPSS screenshots of both the Variable View and Data View, and notes that output highlights are provided. The structure follows a logical methods-section format typical of quantitative research papers.
Multiple regression analysis is a statistical technique used to examine the relationship between one dependent variable and several independent variables simultaneously. In this analysis, the goal is to determine which factors account for the greatest amount of variance in resilience scores among study participants. A hierarchical approach is used, in which variables are entered into the model in sequential blocks so that the unique contribution of each group of predictors can be assessed.
The following variables were entered into SPSS for the analysis:
Demographic variables: Race, Age, Gender, Recruitment Location
Protective factors: Locus of Control, Appraisal, Mentor, Family Mentor, School Experience, Religious Faith
Socioeconomic and adversity variables: Childhood SES, Childhood Maltreatment
Outcome variable: Resilience Composite Score
Each variable represents either the item totals for individual questionnaires included in the survey or responses to individual questions. For example, Race was measured by asking participants to select one of three options. Age was recorded as a numeric entry. Gender was answered by selecting one's gender. For measures such as Locus of Control and Appraisal, responses to multiple items were averaged to produce a single scale score. These variables are then used collectively to conduct the multiple regression.
The hierarchical multiple regression is conducted by entering different variable blocks at different steps in the model. This approach allows the researcher to evaluate how much additional variance in the outcome each new block explains, after accounting for the variables already in the model.
The dependent variable is Resilience, because the study aims to determine which factors best predict resilience among the sample.
The four steps of the hierarchical model are structured as follows:
Step 1 — Demographic variables: Race, Age, Gender, and Recruitment Location are entered first to control for demographic differences among participants.
Step 2 — Protective factors: Locus of Control, Mentor, Family Mentor, School Experience, Religious Faith, and Appraisal are entered as a block. In this study, these variables are conceptualized as protective factors that may buffer against adversity.
Step 3 — Childhood SES: Childhood socioeconomic status is entered to assess its contribution to resilience beyond demographic and protective factor variables.
Step 4 — Childhood Maltreatment: Childhood maltreatment is entered last to determine how much unique variance in resilience it explains after all prior variables have been accounted for.
The results of the analysis indicate to what extent each additional step improves the model's ability to explain variance in resilience scores. While all of these variables may ultimately correlate with resilience — either positively or negatively — not all of them will continue to add predictive value once the contributions of the other variables have already been taken into account. This is one of the primary advantages of the hierarchical approach: it clarifies the incremental variance explained by each predictor block.
You’re 66% through this paper. Sign up to read the remaining 2 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.