A Five Factor Model of Person & Environment Fit

Introduction

Person pushing a puzzle piece into place to symbolize a person environment fit

People’s perception of how well they fit in their environment is a good predictor of satisfaction and duration of participation in that environment (Beasley, Jason, & Miller, 2012). Previous research has indicated that person-environment fit (PE-fit) is best when people perceive they are similar and have congruent values with those in the environment, when their needs are met within the environment, when they are able to meet the demands of the environment, and when they feel they can make a contribution to whatever is happening in the environment. These components were first described in employment settings and more recently, were explored in transitional housing. In this study, we used the same measure with adults who report ongoing impairments to examine the factor structure of PE-Fit in the rural community.

Sample

We recruited a sample of 244 people with impairments from 12 rural communities that represented all four US census regions to complete a survey. More detailed information about the sampling methods and sample characteristics can be found on the Ecology of Rural Participation research project page.

Measures

We used the General Environment Fit Scale (GEFS; Beasley, Jason, & Miller, 2012) for this analyses. Items were adapted to change the wording from transitional housing to community. For example, “The things that I value in life are very similar to the things that my Oxford House values” was changed to “The things that I value in life are very similar to the things that other people in my community value.”

Results

To investigate the factor structure of our data, we used principal component analyses to examine an unrestricted model and a model restricted to either one or five factors using the same 23 items of the GEFS reported by Beasely, Jason & Miller, (2012). For the unrestricted model, six factors had eigen values greater than 1.00, cumulatively accounting for 70.24% of the total variance. In contrast, restricting the model to one factor accounted for 39.65% of the variance. Because prior research indicated the GEFS has five dimensions (listed above) we also computed a five factors solution which accounted for 66.65% of the total variance. The structure matrix is shown below in Table 1 and the component matrix is shown in Table 2.

Table 1

Five Factor Model Table 1Table 2 

Five Factor Model Table 2