week 7

1

5

Writing the First Section of the Criminal Justice Proposal

First Section of the Criminal Justice Research Proposal

Research question: To what extent is residents’ exposure to or perceptions of community-policing practices associated with their levels of trust in law enforcement and perceptions of neighborhood safety in Miami, Florida?

Subproblems

One of the main research challenges is the lack of research evidence concerning the link between community-policing practices, residents' trust in police, and perceptions of neighborhood safety. The subproblems are: (1) What is the relationship between residents' perception of the role of police visibility and their level of trust in the police? How does residents' trust relate to how they perceive that the police are doing their job fairly and promptly? (3) What is the relationship between community involvement in policing and people's notions of neighborhood safety?

Theoretical or Conceptual Framework

The research will adopt procedural justice theory for guidance. The theory suggests that an individual's feelings of fairness in their encounter with authority affect their trust, legitimacy evaluation, and cooperation with authority. In the context of community policing, residents feeling fair, respectful, transparent, and responsive to citizens might be more likely to trust policing and cooperate with law enforcement (Hevi et al., 2022).

A Priori Hypotheses

The first hypothesis states that residents who report greater perceptions of fairness and responsiveness in community-policing practices will report higher levels of trust in law enforcement. The second hypothesis states that greater exposure to or perceptions of community-policing practices will be positively associated with residents’ perceptions of neighborhood safety. These hypotheses are theoretically sound and have been backed up by research that supports the idea that a well-designed community-policing model might help to promote safety and crime prevention in the neighborhood (Alehegn et al., 2025).

Key Concepts and Variables

The independent variables are; police visibility, communication, fairness, being attuned, and community participation. They will be evaluated by residents' responses to a survey asking them about their experiences and perceptions of the relationship with the police. Perceived neighborhood safety and trust in law enforcement will be considered dependent variables.

My Assumptions

The study takes the stance that participants will remain honest with the researcher in their answers with regard to experiences and perceptions of the police. It also assumes the questions in the survey will be comprehensible to respondents, and the respondents will have some familiarity with or awareness of the policing practices in their community, or local neighborhood, in order to answer the questions.

Delimitations

It will include only community members, and the relationship among community-policing practices, trust in the police, and neighborhood safety will be examined. It will not review other areas of the police service like the salaries of police officers, various criminal sentencing options, recruitment for officers, or national crime patterns.

Possible Limitations

Some limitations are self-report bias, as the participant might give socially desirable answers or their personal experiences may affect the answers. Some of the limitations of sampling may also limit the generalizability of findings (Ormrod, 2023).

References

Alehegn, D., Karunakaran, R., & Engida, B. (2025). The role of community policing structures in enhancing safety, security, and local crime prevention in Addis Ababa City Administration. SN Social Sciences, 5(10), 168.

Hevi, S. S., Ketemepi, G. E., Dorkoo, C., & Wuttor, A. (2022). Community policing experience, public trust in the police, citizens’ psychological safety and community well-being in Ghana. Safer Communities, 21(2), 123–136.

Ormrod, J. E. (2023). Practical research: Design and process (13th ed.). Pearson Education.

Planning a Quantitative Data Analysis

Characteristics of the Data

1. For each variable, are the data continuous or discrete?

The main variables to be studied will be considered continuous variables for the generation of composite scores using several Likert scale questions. These variables include police visibility, communication, procedural justice, responsiveness, community involvement, trust in law enforcement, and neighborhood safety (Ormrod, 2023). The demographic variables such as gender and education will be considered discrete variables.

2. For each variable, what scale do the data reflect? Are they nominal, ordinal, interval, or ratio?

Responses to individual Likert-type questions are ordinal, since they reflect varying degrees of agreement. Composite scales based on several questions will be treated as roughly interval level (Hennigar et al., 2024). Variables like gender that describe categories will be nominal. Age and time living in the area are ratio-level variables.

3. For each variable, are the data likely to approximate a normal distribution?

Yes. The composite scores are supposed to follow normal distribution, at least with an adequate sample size. These histograms, Q-Q plots, and skewness/kurtosis tests will be analyzed before the inferential analysis is performed.

4. What do you want to do with the data?

Mean values and median will describe measures of central tendency, while standard deviation and range values will be measures of dispersion. The relationship between community policing activities, trust in the police, and the perception of neighborhood safety will be determined using Pearson correlation coefficients (Ormrod, 2023). The null hypothesis will be tested using a confidence interval of 95% with an alpha value of .05. Multiple regression analysis will also be carried out to test if there is any correlation between the predictors.

Rationale for Processing the Data

5. State your rationale for processing the data as you have just said you intend to do.

The above methodologies would be relevant since the study will utilize survey data based on quantitative research design in order to investigate the correlations between community policing practices, level of public trust, and safety perceptions. The descriptive statistics will help in analyzing the sample size as well as the major variables of the research. The Pearson correlation methodology will establish the nature and strength of correlations while multiple regressions will test the relationships between several predictors and the outcome variables. The data will also be analyzed for outliers, missing values, normality, and other assumptions.

Interpretation of the Data

6. After you have statistically analyzed your data, what information will you then have?

This will help to give insight into how people view such aspects of policing like visibility, communication, fairness, responsiveness, community involvement, trustworthiness of law enforcers, and safety within neighborhoods (Miola & Miot, 2022). The results from correlation will indicate the relationship between the variables while those from regression analysis will reveal which factors influence trust and safety in neighborhoods.

7. From a research standpoint, what will your interpretation of the data entail? How will the results of your statistical analyses help you address some or all of your research problems or questions?

In order to interpret the findings, it will need to be determined if the findings are in accordance with the research questions and hypotheses. For instance, if there exists a statistically significant positive association between procedural justice and trust, then it will imply that high levels of perceptions of fairness are linked to high levels of trust.

8. Will some of your data be inappropriate for statistical analysis? If so, how might you analyze and interpret them?

Yes. It would be wrong to use open-ended questions in correlational or regression analysis. However, they can be analyzed using thematic and content analysis to discover recurring themes. It is possible to code categorical variables for regression analysis. Missing values and outliers will also be checked before statistical analysis. Thus, the use of descriptive and inferential statistics will help analyze the phenomenon systematically.

Reference

Hennigar, A., Piatt, R., Alves, S., & Friend, D. (2024). Cleaning and preparing data for analysis. HMRE HMRE, 31.

Miola, A. C., & Miot, H. A. (2022). Comparing categorical variables in clinical and experimental studies. Jornal Vascular Brasileiro, 21, e20210225.

Ormrod, J. E. (2023). Practical Research: Design and Process (13th ed.). Pearson Education (US). https://ccis.vitalsource.com/books/9780138088781

Specific feedback for assignment 4

Travis, this is a strong quantitative analysis plan that demonstrates continued progress in translating your research problem into measurable variables and appropriate statistical procedures. You have moved beyond simply stating that your study will be “quantitative.” You are beginning to explain what will be measured, how the data will be classified, what statistical procedures will be used, and what those procedures can tell you about your research question. One of the strongest portions of your assignment is your treatment of measurement levels. You correctly recognize that an individual Likert-type response is ordinal while a composite score constructed from multiple items may, under appropriate conditions, be treated as approximately interval-level for analysis. You also correctly distinguish nominal demographic variables from ratio-level measures such as age and length of residence.

Your response to Question 3 also demonstrates good methodological awareness, as you do not simply assume normality. You indicate that histograms, Q-Q plots, and skewness/kurtosis would be examined before conducting inferential analyses. This is precisely the kind of planning the exercise is intended to encourage.

Statistical Analysis

Your proposed use of descriptive statistics, Pearson correlation, and multiple regression is generally logical for the study you are developing. You propose means and medians for central tendency, standard deviations and ranges for variability, Pearson correlations to examine relationships among community-policing measures, trust, and perceived safety, and multiple regression for a more developed analysis. There is one important distinction to clarify.

You state that:

“Multiple regression analysis will also be carried out to test if there is any correlation between the predictors.”

That is not quite the purpose of multiple regression. Pearson correlation can estimate the direction and strength of a bivariate linear relationship between two continuous variables. Multiple regression, by contrast, allows you to estimate the relationship between multiple predictor variables and a specified dependent variable while statistically accounting for the other predictors in the model.

For example, your eventual model might examine whether:

Police visibility + communication + procedural justice + responsiveness + community involvement → trust in law enforcement

A separate model could examine whether those variables—and perhaps trust itself—are associated with:

Perceived neighborhood safety

This distinction will become important when you formally identify your independent and dependent variables.

Be Careful With Causal Language

You write that regression will reveal which factors “influence trust and safety in neighborhoods.”

Because your proposed study appears to use a cross-sectional survey rather than an experimental design, regression alone cannot establish that one variable caused another. A statistically significant regression coefficient could demonstrate an association or predictive relationship, but not necessarily causation.

A more precise statement would be:

Regression analysis will identify which community-policing variables are significantly associated with, or statistically predict, residents' trust in law enforcement and perceived neighborhood safety after accounting for the other variables included in the model.

That distinction is important at the graduate research level.

Normality Requires Some Additional Precision

Your statement that composite scores are “supposed to follow normal distribution” should also be modified. An adequate sample size does not guarantee that every variable will be normally distributed. Your proposed diagnostic procedures—histograms, Q-Q plots, skewness, and kurtosis—are therefore appropriate.

Also remember that the relevant statistical assumptions depend on the particular procedure being used. When you reach regression analysis, you will need to consider assumptions involving linearity, residuals, homoscedasticity, independence, influential observations, and multicollinearity, rather than focusing only on whether each raw variable appears normally distributed.

Question 8 Is Particularly Strong

Your response regarding data that are inappropriate for statistical analysis demonstrates good research design thinking. You recognize that open-ended survey responses should not simply be inserted into Pearson correlations or regression models and instead propose thematic/content analysis. You also recognize that categorical variables can sometimes be appropriately coded for quantitative analysis.

One refinement: missing values and outliers are not necessarily “data inappropriate for statistical analysis.” They are data-quality and diagnostic issues requiring evaluation before analysis. Depending upon their nature and extent, researchers must determine an appropriate strategy for handling them rather than automatically excluding them.

APA and Presentation

The reference section contains three sources, satisfying the rubric's research requirement, but APA formatting needs another review. Travis Mccalman_Writing Assignm… The heading should be References, the textbook entry should follow APA 7 book-reference conventions, and the Hennigar et al. citation should be carefully checked for complete and accurate publication information. On the title page, change “September 28,2026” to “September 28, 2026.”

Overall, this assignment demonstrates strong alignment among your research question, variables, measurement strategy, and proposed statistical analysis. Your principal task now is to sharpen the distinction between correlation and regression, avoid causal interpretations that the proposed survey design cannot support, and more precisely specify which variables will serve as predictors, and which will serve as dependent outcomes in each regression model.

Final Score: 46/50 — 92%