Skip to main navigationSkip to main content
The University of Southampton
Practical Applications of Statistics in the Social Sciences

Research Question 1: Confidence in the police

Our research question for the whole of this section is: What factors influence confidence in the police?

Suppose we are interested in improving the public’s confidence in the police in their neighbourhoods. Policy makers would like residents to see the police as capable and competent, and as an effective means by which crime is reduced and public safety is achieved. We want to know what individual characteristics may impact a person’s perception of the police and their confidence in police, because this data can inform policies aimed at improving public confidence in policing.

This question can be approached in several different ways. We could conduct qualitative research, employing techniques like ethnography, observation, or interviews to investigate how people in various neighbourhoods feel about the local police. Alternatively, we can use quantitative methods like statistical analyses of collected survey data to tease out trends and differences in levels of police confidence. Qualitative and quantitative research can be complementary and used to serve the same purpose. For example, you could use qualitative observation to identify areas of analysis and then quantitative methods to determine whether any statistically significant relationships exist in your data.

If you do decide to conduct quantitative research, you can collect your own survey data or use data that has already been collected. A major benefit to using data that has already been collected is that it usually is a large sample which is representative of the whole country from which it was taken.

In addition, the UK Data Service hosts over 6,000 datasets which are publicly available, which makes finding and selecting data to analyse very convenient. For our analyses here, we’ll be using a dataset available on the UK Data Service website. The dataset we’ll be using is the csew_apr11mar12_nvf.sav file from the Crime Survey for England and Wales (CSEW) in 2011-2012. You can find out how to locate, download, and access the CSEW from the UK Data Service website here.

Before we can begin using this dataset to investigate the possible answers to our research question, we need to identify which variables we’ll be using. This process is sometimes called the operationalization of concepts; simply put, we need to decide what ideas we are interested in and then determine which of our available variables allow us to define those ideas.

In our case, because we are interested in determining what factors influence confidence in the police, we will operationalize the level of confidence the respondents have in the police by selecting a variable that concerns the survey respondents’ opinions about the police. This will be our dependent variable, as it is the variable that we think might be influenced by other respondent characteristics, or independent variables.

Throughout the analyses we run in this section, we’ll be using different independent variables to illuminate relationships between our police confidence dependent variable and various survey respondent characteristics. Some of these independent variables include respondent health, education, ethnicity, and employment.

As you move through the pages in this section, you’ll learn how to perform various statistical analyses on the data in the CSEW dataset. First, you’ll use univariate analysis to get some descriptive information about our dependent variable. Then, you can run bivariate analysis, introducing an independent variable into statistical tests to define its relationship with police confidence. Finally, you will practice multivariate analysis, which allows you to look at the influence of multiple independent variables on a dependent variable.

Privacy Settings