Demography and Geography in Performance Measurement: Exploratory Analysis

Counties act as the state's agents for delivering human services to residents, utilizing federal and state funds to fulfill this role. The state supervises the delivery system by evaluating performance. The county administrator closely manages the system, concentrating on specific mandated measures. To comply with these mandates, the administrator aims to comprehend demographic and geographic factors that are frequently outside management's control. This article will demonstrate an exploratory approach to understanding the impact of external factors on performance measurement.

Minnesota’s Human Services Performance Management System (1) Measure 1, under target Outcome 1, states: “Percent of children with a substantiated maltreatment report who do not experience a repeat substantiated maltreatment report within 12 months. (2)” The benchmark is set at 90.9%. Penalties apply for not achieving this benchmark, the county administrator to identify external barriers to the defect-free delivery of human services to Minnesota residents.

A correlation matrix was employed to examine the relationships between pairs of measures and factors. An increase in one factor is positively associated with an increase in another. The analysis indicated that adjustments are needed for measured performance to consider the factors of “American Indian/Alaskan,” “Asian,” and “population per square mile.”

Methodology

In this analysis, the term "number of cases" refers to the count of children with a confirmed maltreatment report, which acts as the measurement base. The "Standard" is defined as 90.9% of this base, while the "Measure" represents the actual number of cases excluding repeat child maltreatment reports. "Performance" is calculated by subtracting the Standard from the Measure.

A “flat” Excel file containing 2020 measure and standard data for each of Minnesota’s 87 counties was obtained from the Human Services Performance Management System upon request. Data from 52 counties were deemed usable for the analysis. Twelve demographic data points and one geographic data point for each county were downloaded from the website of the Association of Minnesota Counties (3). The data were organized into columns in an Excel worksheet in preparation for analysis. Accessing the correlation tool from the worksheet main menu involved navigating to: Data > Data Analysis > Correlation (4).

Observations

The Correlation Matrix (4) that was developed produced the following observations:

1. The case data has a 95% correlation with the population data. Conducting a simple linear regression of these two data sets resulted in an average of 1.2 cases per 1,000 residents in the county (5)(6).

2. The data for “American Indians/Alaskans” and “Asians”show the lowest correlation with the number of cases at 70% and 89%, respectively.

3. The geographic factor “population per square mile” shows a lower correlation with the number of cases data at 78%.

4. The standard accounts for 90.9% of a county’s total cases.

5. The county's performance data (measured – standard), which shows a 74% correlation, together with the number of cases, reveals how effectively a county’s delivery process can address process defects and environmental factors.

6. In this context, the correlation between the “American Indian/Alaskan” data and performance data is weaker compared to other racial categories, indicating a less strong correlation with the number of cases data.

7. “Asians” data shows a high correlation with the performance data.

8. The population density data for the counties is 88% correlated with the performance data.

9. The “median household income” data does not appear to be connected to the base data or the performance data.

Conclusion

To enable a Minnesota county to meet or exceed the state’s Human Services PerformanceMeasure 1 (Outcome 1) stated as, “Percent of children with a substantiated maltreatment report who do not experience a repeat substantiated maltreatment report within 12 months”, adjustments must be made on measured performance to take into account “American Indian/Alaskan”, “Asian” and “population/square mile” factors (7).

By Noel Jagolino, Management Consultant

    Contributed article to Www.mgmtlaboratory.com 2025.

References:

(1) A section currently within the Minnesota Department of Children, Youth, and Family.

(2) This is Measure 1 under human services target Outcome 1 in 6 outcome categories enumerated in the following table:

a. Outcome 1: Adults and Children are safe and secure. Measure 1 and Measure 2.

b. Outcome 2: Children have stability in their living situation. Measure 1 and Measure 2.

c. Outcome 3. Children have the opportunity to develop to their fullest potential. Measure 1 and Measure 2.

d. Outcome 4. People are economically secure. Measure 1, Measure 2, Measure 3 and Measure 4.

e. Outcome 5. Adults live with dignity, autonomy, and choice. (No measures yet.)

f. Outcome 6. People have access to healthcare and receive effective services. (No measures yet.)

(3) https://www.mncounties.org/information_and_jobs/county_data.php

(4) The correlation matrix was developed with Microsoft Excel using this procedure: Data > Data Analysis > Correlation.

(5) SLRA: A Multi-Purpose and Accessible Executive Analytics Tool. www.mgmtlaboratory.com/blog. May, 2018.

(6)

(7) Evaluating Human Services Performance across Counties: A Multiple Regression Analysis (MRA) Demonstration. www.mgmtlaboratory.com/blog. Jul, 2018.

Mgmtlaboratory.com staff and affiliated management consultants are experts in managing private and public organizations and in using managerial tools. Government entities can contact contact@mgmtlaboratory.com to inquire about its free online consulting service on continuous improvement and other management tools.