B/CS Habitat for Humanity
Neighborhood Impact Study

August 9, 2012

Researcher

Qian “Eric” Luo

Candidate of Master in Public Service and Administration

George Bush School of Government and Public Service

Texas A&M University

Graduate Intern – Summer 2012 at B/CS Habitat for Humanity

Contents

Crime Rate Impact Study

Abstract

Executive Summary

Key Findings

Methods and Indicators

Results

Conclusion

Reference

EconomicImpactStudy (ForCalendarYear2009-2011)

Executive Summary

Results

Value in Building Activities

Value in Volunteer Activities

Tax Value

Savings to Habitat homeowners per month

Homeowner Survey Report

Preface

Executive Summary

Key findings

Introduction

Methods

Result

Limitations and Future Research:

Appendix A – Sample Questionnaire

Appendix B – Table

Appendix C – Graphs

Bryan/College Station Habitat for Humanity

Neighborhood Impact Study of 2012

Crime Rate Impact Study

Abstract

B/CS Habitat for Humanity has built more than 200 houses in Bryan/College Station area. It has impacts on not only the homeowner families but also the local community. In this report, we have primarily tested three hypotheses about the impacts of Habitat subdivisions on local community crime rate: 1) in-process subdivisions can lower the crime rate in the neighborhood; 2) completed subdivisions has a lower-than-average crime rate; 3) in-process subdivisions can lower the crime rate in the nearby area. According to our findings, hypotheses 1) and 3) are supported, whereas hypothesis 2) is only partially supported. Overall, our findings support that Habitat subdivisions have positive impacts on lowering the local crime rate.

1

Bryan/College Station Habitat for Humanity

Neighborhood Impact Study of 2012

Executive Summary

Habitat for Humanity International, founded in 1976, is an ecumenical Christian housing ministry that has built over 500,000 homes and served about 2.5 million people across the globe. Bryan/College Station Habitat for Humanity (B/CS HfH), a Texas based affiliate of Habitat International, has served the local region since 1989.

The mission of all Habitats is to eliminate inadequate and substandard housing through the construction of decent quality homes. Through a proven selection process, B/CS HfH welcomes qualified families to become partners towards the achievement of this mission. Families graduate from home ownership curriculum and earn at least 500 sweat equity hours: B/CS HfH homes are not given; they are earned.

Habitat B/CS has a business model that transforms both the lives of the family and the broader community; one of those positive observed impacts is the decrease of the local crime rate. This study considers the impact of three major subdivisions built by B/CS Habitat for Humanity— Angel’s Gate, Miracle Place, and Sharon’s Court. Primarily, we expect three possible impacts on the crime rate:

1)The under-construction subdivision (Angel’s Gate) will lower the crime rate.

2)The completed subdivision (Miracle Place & Sharon’s Court) will be a safer place than the average of the City of Bryan, which means that the crime rate would be consistently lower than the average of the City of Bryan.

3)The under-construction subdivision (Angel’s Gate) has positive externalities on nearby area, including the Sharon’s Court subdivision.

To test the three above hypotheses, the City of Bryan Police Department has shown great support through the provision of data. Thanks to Bryan PD the researcher has access to annual data from 2006-2010. With this information, the researcher can conduct a basic statistical study to analyze the impact of our subdivision on the local crime rate.

Key Findings

According to the results, our hypotheses are partially supported: the under-construction subdivision has some impact on the decrease of crime rate, but there is no statistically significance associated with the conclusion that our completed subdivision can contribute to a lower-than-average crime rate in the City of Bryan.

  • Hypothesis I. The in-process subdivision (Angel’s Gate) built by B/CS HfH lowers the crime rate in local neighborhoods.
  • Hypothesis II. On average, the completed subdivision (Miracle Place) built by B/CS HfH has a lower crime rate than the City of Bryan. Although there is barely statistical significance for this conclusion, we can expect a more solid conclusion from more data under the same pattern in the future.
  • Hypothesis III. The in-process subdivision (Angel’s Gate) also lowers the crime rate in the nearby areas.

Methods and Indicators

Primarily, we employ the descriptive methods to analyze the data due to the limitations; nonetheless, we also attempt to build some pilot inferential models for the crime rate in our subdivision.

According to a previous study by the Harries County Habitat affiliate (2006), the crime in an area is associated with the population density in the geographic area. However, when we compare crimes between areas with dramatically different area, the crime rate determined by population density could be dramatically different from each other, since we forget to take into account the acreage of the area. Thus, for the indicators, we employ the commonly used crime rate indicator crime-per-person. Although the other study may use the crime per one hundred thousand as the indicator, it does not suit for our situation with neighborhoods for less than 500 hundred people.

Results

Hypothesis I

  • Hypothesis I states that the in-process subdivision (Angel’s Gate) built by B/CS HfH lowers the crime rate in local neighborhoods. Figure 1 shows the data for the crime rate in Angel’s Gate, Miracle Place, Sharon’s Court and the City of Bryan.

Figure 1 Crime Rate Comparison

From the data, we can see that the crime rate in Angel’s Gate drops dramatically since it started in 2006. Further, although the fluctuation of the crime rate in Miracle Place and Sharon’s Court is somewhat large, we can expect a lower average crime rate in five years than the City of Bryan.

Figure 2 Comparison between Angel’s Gate and the average of the City of Bryan

In Figure 2, a closer examination shows the positive impact of the in-process subdivision on the crime rate. We analyze the curve of the crime rate data in the City of Bryan and Angel’s Gate subdivision and use OLS method to estimate the change of the data. Table 1 shows our estimations of these two curves.

Table 1 Regression Result

Area / Equation / Significance of Slope (P-value) / R2
Bryan / / 0.481 / 0.1766
Angel’s Gate / / 0.013** / 0.9030
Note: ** means significant at 5% level.

The regression result shows that the slope in the City of Bryan is not statistically significant (0.481>0.05), which means that the slope should be zero. Therefore, we can draw the conclusion that the crime rate in Bryan during our studying period is stable. The result also shows that the slope in Angel’s Gate subdivision is highly significant (0.013<0.05). Thus, we confirm that the Crime Rate in Angel’s Gate went down by 0.019 per year during the studying period. The R-squared is the goodness of fit, which indicates how well the model fits the data. In Angel’s Gate, the R-squared is 0.9030, which means that 90.3% of the change in crime rate in Angel’s Gate is explained by our model.

However, the simple result of crime rate decrease by 0.0186 per year does not mean much for those who have not taken a statistical course. Thus, we switch to a semi-log model that shows the percentage points change in crime rate.

Figure 3 Semi-Log Model of Change in Crime Rate

Figure 3 shows the graph of semi-log model, and Table 2 presents the results of semi-log estimation.

Table 2 Semi-Log Regression Result

Area / Equation / Significance of Slope (P-value) / R2
Bryan / / 0.461 / 0.1916
Angel’s Gate / / 0.022** / 0.8658
Note: ** means significant at 5% level.

Under a semi-log model, we can interpret the slope in a more understandable way. In this model, our slope is also statistically significant in Angel’s Gate. The slope means that in Angel’s Gate, the crime rate decreases by 30.1 percentage points per year during our studying period. (Notice: Here the change is in “percentage points” rather than “percent”.)

Hypothesis II

As for the comparison of completed subdivision and the City of Bryan, we find only partial support for our second hypothesis. As Figure 4 shows, the average crime rates from 2006 to 2010 in Miracle Place and Sharon’s Court are lower than the average of the City of Bryan.

Figure 4 Comparison between Miracle Place and the City of Bryan

This appears to support our second hypothesis, but a further examination for those data does not prove our hypothesis. We use the Student’s T-test to determine whether the difference between the average crime rate during the studying period is statistically significant or not. If the p-value of t-test is smaller than 0.05, we would accept it as highly statistically significant. If the p-value of t-test is smaller than 0.10, we would accept it as statistically significant. Otherwise, we would not. Table 3 shows the results for t-test.

Table 3 Results for the Student’s T-test.

Hypothesis / T-test Result (p-value)
/ 0.1145
/ 0.1464

The t-test shows that none of two completed subdivision is significantly different from the average of the City of Bryan. In other words, we do not have enough evidence to prove that the completed subdivision has a lower-than-average crime rate. Nonetheless, the p-values for our hypotheses are not too large, and usually we can say that they are barely significant. If there are more data in the similar pattern during the studying period in the future, we can expect a significant result that confirms our hypotheses.

Hypothesis III

As the foregoing discussion mentioned, we expect positive externalities on the areas around our in-process subdivisions. To test this hypothesis, we conduct a crime rate analysis for Sharon’s Court subdivision and Angel’s Gate subdivision, since they are geographically adjacent.

Figure 5 Crime Rate Curves

Figure 5 shows the crime rate curves of Sharon’s Court and Angel’s Gate. From the graph, we can see the trend line is somewhat downward. To get more convincing evidence, we perform a regression for those two areas. Table 4 presents the results.

Table 4 Regression Result

Area / Equation / Significance of Slope (P-value) / R2
Sharon’s Court / / 0.262 / 0.3871
Angel’s Gate / / 0.013** / 0.9030
Note: ** means significant at 5% level.

In the results, Sharon’s Court shows somewhat downward trend, but it is not statistically significant. Although there is no significance associated with data from Sharon’s Court, it is still acceptable since Sharon’s court is not the only neighborhood around the Angel’s Gate. Thus, to perform a further analysis, we can study Sharon’s Court and Angel’s Gate as a whole. If the result is associated with better significance with slope and better goodness of fit than solely analyzing the Angel’s Gate, we can draw some conclusions from there.

Figure 6 shows the curve of the crime rate in A&S area (Angel’s Gate and Sharon’s Court).

Figure 6 Crime Rate Curve in A&S Area

This graph shows that the crime rate in A&S area is significantly downward. Therefore, we perform a regression analysis to confirm our result. Table 5 shows the regression result for A&S area. The slope is highly significant (0.006<0.05), and it is better than the result for Angel’s Gate. The goodness of fit is also better than the result for solely Angel’s Gate (0.9392>0.9030). This indicates that the A&S area went down with the progress in Angel’s Gate subdivision, and, therefore, it confirms the hypothesis that in-progress subdivision can also lower the crime rate in nearby areas.

Table 5 Regression Result

Area / Equation / Significance of Slope (P-value) / R2
A&S / / 0.006** / 0.9392
Note: ** means significant at 5% level.

To make the result more readable, we use the semi-log model as the foregoing discussion. Table 6 shows the result for the semi-log regression.

Table 6 Semi-Log Regression Result

Area / Equation / Significance of Slope (P-value) / R2
A&S / / 0.006** / 0.9437
Note: ** means significant at 5% level.

The result represents that the crime rate in A&S area went down by 17.2 percentage points per year during the studying period. This model also explains the variation in crime rate better than solely analyzing the Angel’s Gate (0.9437>0.8658).

Conclusion

Based on the foregoing discussion, we can draw the following conclusion:

1)Angel’s Gate does lower the crime rate as an in-process subdivision.

2)Miracle Place and Sharon’s Court show a lower crime rate than the city average; however, the difference is not quite statistically significant. Nonetheless, we can expect a more solid result with more data in the pattern during the studying period.

3)Angel’s Gate does have positive externalities to lower the crime rate in the nearby areas, such as the Sharon’s Court.

Overall, this study suggests that Habitat subdivisions have some positive impacts on the local crime rate. B/CS Habitat for Humanity can lower the crime rate in the area which initially has a higher crime rate, but no more evidence shows that the completed subdivisions have a lower-than-average crime rate.

Reference

Harries, Keith. 2006. Property Crimes and Violence in United States: An Analysis of the influence of Population density. International Journal of Criminal Justice Sciences Vol. 1 Issue 2 July 2006,Accessed: June 18, 2012 <

Phillips, Iris, Bennett, Stephanie, Opatrny, Marie, Priest, Ronda and Khayum, Mohammed. 2011. Habitat for Humanity Impact Study forHabitat for Humanity of Evansville.

Bryan Police Department. 2007. Bryan Uniform Crime Reporting of 2006.

Bryan Police Department. 2008. Bryan Uniform Crime Reporting of 2007.

Bryan Police Department. 2009. Bryan Uniform Crime Reporting of 2008.

Bryan Police Department. 2010. Bryan Uniform Crime Reporting of 2009.

Bryan Police Department. 2011. Bryan Uniform Crime Reporting of 2010.

Economic Impact Study (For Calendar Year 2009-2011)

Executive Summary

This study is a snapshot of the economic impact of Bryan/College Station Habitat for Humanity (B/CS HfH). Based on the document “Determining Your Affiliate’s Economic Impact- a Formula for Success”, we analyze the economic impact of our summary from the calendar year 2009 to 2011.

Results

Value in Building Activities

  1. From 2009 to 2011, B/CS HfH built 40 houses. The average sales price is $72,347.50.

This is the total sales revenue we generate in the community.

In the City of Bryan, the average sales price is $72,666.67.

In the City of College Station, the average sales price is $69,475.00.

  1. The average cost of these building activities is

Thus, using the 7 multiplier, the total economic impact of these building activities is

  1. The average assessed value (appraised value) of these houses is $67,691.50.

In the City of Bryan the average appraised value is $65,470.83. On average, the appraised value is 10% lower than its sales price.

In the City of College Station the average appraised value is $87,677.50. On average, the appraised value is 28.11% higher than its sales price.

Value in Volunteer Activities

The following table shows the volunteer hour value in Texas from 2009 to 2011.

Year / Volunteer Value in Texas / Volunteer Value
2009 / $21.35 / $20.85
2010 / $21.91 / $21.36
2011 / $22.54 / $21.97

The following table presents the dollar value of our volunteer activities.

Sweet Equity / Other Volunteer Hours / Total
Year / Hours / $ Value / Hours / $ Value / Hours / $ Value
2009 / 10236.8 / $218,555.68 / 21737 / $464,084.95 / 31973.8 / $682,640.63
2010 / 9114.8 / $199,705.27 / 19306 / $422,994.46 / 28420.8 / $622,699.73
2011 / 9479.6 / $213,670.18 / 18953 / $427,200.62 / 28432.6 / $640,870.80
Sum / $631,931.13 / $1,314,280.03 / $1,946,211.16

Tax Value

Year
Entity / 2009 / 2010 / 2011 / Sum
City of Bryan / $58,404.03 / $64,971.44 / $69,831.29 / $193,206.76
Bryan ISD / $95,629.46 / $107,852.90 / $115,449.99 / $318,932.35
City of College Station / $6,324.50 / $6,801.85 / $7,560.00 / $20,686.35
College Station ISD / $14,563.46 / $16,404.49 / $19,645.60 / $50,613.55
Brazos County / $49,674.40 / $55,570.03 / $59,567.11 / $164,811.54
Sum / $224,595.85 / $251,600.71 / $272,053.99 / $748,250.55

Savings to Habitat homeowners per month

Bryan / College Station
Median Gross Rent / $748.00 / $833.00
Median Monthly Payment / $353.18 / $369.73
Monthly savings / $394.82 / $463.27

Homeowner Survey Report

Preface

This homeowner survey is a component of the impact study conducted by the Bryan/College Station Habitat for Humanity. The study consists of three separate studies that focus on the impact of the Habitat houses to the local economy, local crime rate and the homeowners. The author is a current student at the Bush School of Government and Public Service, and 2012 summer intern at Habitat for Humanity. The executive director- Marco Maine, the director of volunteer and communication- Ryan Pierce, and my supervisor Leah Morales provide great help during the conduct of this research.

Executive Summary

Bryan/College Station Habitat for Humanity is a nonprofit organization located in the City of Bryan, Texas with a mission focused on eliminating poverty housing and creating better neighborhoods and communities. B/CS Habitat for Humanity brings carefully selected families into the local neighborhood and community. Through its home building activities, B/CS Habitat for Humanity impacts the lives of the homeowner families.

As the foregoing paragraph discussed, this study considers the impact of the Habitat houses on more than two hundred homeowner families. We mailed out the survey to 198 homeowner families and got 21 respondents. The response rate is 10.6%. This response rate is problematic, and it may undermine the reliability of this study. Nevertheless, we conduct this study as a preliminary study; thus, there is still some information we can get from the survey result. Moreover, this study can provide some guidance for further impact study in the future.

According to our survey result, we find that having a habitat house has positive impacts on the homeowner families in economic, psychological, social, and educational aspects. Meanwhile, the mortgage payments somewhat increase the financial obligation of the families, although the difference is not statistically significant.

Key findings

Economic Aspect:

  • Homeowner families report significantly more positive regarding financial status. They reported:
  • … increase in their influence on what happens to them.
  • … adults have made work-related achievements.
  • Homeowner families did not perceive financially better off in the aspects of paying bill. They reported:
  • … no statistically significant difference in ability to pay the bills.

Psychological Aspect:

  • Homeowner families report more positive regarding the emotional status and mastery on their life. They reported:
  • … increase in control on their life.
  • … decrease in emotional stress among family members.
  • … seems to be happier than many other families they know.
  • Homeowner families did not report improvement regarding the esteem. They reported:
  • … no statistically significant difference in their ability to “look at the brighter side of things.”
  • … no statistically significant difference in the degree to which the family members respect each other.

Social Aspect:

  • Homeowner families report more positive regarding the acceptance by their community and better community atmosphere. They reported:
  • … increase in involvement with the neighborhood activities.
  • … increase in frequency of attending churches.
  • … racial harmony is more positive (marginally significant).
  • Homeowner families report safer in a habitat house/subdivision. They reported:
  • … crime rate is lower than their previous communities.
  • … drug use/dealing is lower than their previous communities.

Educational Aspect:

  • Homeowner families report that children in the families gain more educational achievements after having a Habitat house, but there is not enough evidence about whether the adults in the family gains more educational achievements than before.

Introduction