Exercise 3: Analysis and Conclusion
Finding and Analyzing Data
Directions: Students will complete a multi-step activity to learn how to find raw datasets, explore data using a statistical software program, and report the findings of data. Students will complete an exercise to learn how to conduct statistical analysis for research. Students will practice writing the results section of a research project. This will not be the analysis and findings section in your research project but rather an exercise to learn how you would go about completing it. To complete the exercise, complete the following steps.
Step 1: Downloading SPSS
SPSS is a statistical software program that aids researchers in recording and analyzing data to make interpretations. In previous modules you learned about the types of statistics: descriptive and inferential. Descriptive statistics are used to describe data. This allows us to understand our sample, and what the data can tell us. Inferential statistics allows us to make interpretations about our data. It aids us in answering our research questions and either rejecting or accepting the null hypothesis.
SPSS will make our job easier. It does all the math for us! We do, however, have to know what the outputs it gives us, mean. To complete this step, you will need to download SPSS from the USF IT website: here.
Before you download it, be sure you know whether you have a PC or Mac and what type of operating system you have. If you need assistance, please reach out to the IT help desk and they will assist you. You can also stop by. This is a very expensive software that is provided free to all students. It is important that you download it rather than accessing it online. Once you have it downloaded, you will be able to access the data that you have downloaded.
Step 2: Finding Raw Data
Once you have downloaded SPSS, you will need to find some raw data to work with. There are many resources for getting raw data, especially Criminal Justice data. The following websites offer criminal justice data for download in SPSS:
1. ICPSR – https://www.icpsr.umich.edu/web/ICPSR/search/studies?start=0&ARCHIVE=ICPSR&PUBLISH_STATUS=PUBLISHED&sort=score%20desc%2CTITLE_SORT%20asc&rows=50&q=
2. GSS – https://gss.norc.org/get-the-data/spss
3. NACJD – https://www.icpsr.umich.edu/web/NACJD/search/series?q=&start=0&sort=TITLE_SORT%20asc&ARCHIVE=NACJD&ARCHIVE=NACJD&ARCHIVE=NACJD&rows=50
ICPSR is the Inter-University Consortium for Political and Social Research and houses over 10,000 datasets. Some of these datasets contain data from protected populations which requires approval for access. However, most of the data is available for download using SPSS. You will need to create an account through the institution using your USF email address.
General Social Survey (GSS) is a survey that has been collected by the University of Chicago since 1972. It collects information on political beliefs, social attitudes, and cultural practices. It is offered by year or as the full dataset.
NACJD is the National Archive of Criminal Justice Data that offers criminal justice specific datasets. Some may be protected but most are available for download using SPSS. They can also be found within ICPSR.
To complete this step, look through each resource to familiarize yourself with the layout and function of their website. Look through the datasets and find one that looks interesting to you and download the data in SPSS.
Step 3: Frequencies, Descriptives, and Measures of Central Tendency
Don’t be overwhelmed with data and numbers! We are going to explore some sample data and run a few descriptive statistics tests using SPSS to begin to understand how to work with datasets and numbers.
To begin, open SPSS. A large window will open and a small pop-up box will open. On the pop-up box, towards the bottom of that box you will see three tabs that say restore points, recent data and sample data. Click on sample data. In the window you will see a list of data files. Scroll until you find recidivism. There are more than one that start with recidivism. Select the one that just says recidivism and click open.
There are two different views for the data: Data view and Variable View. If you look to the bottom left corner, you will see the two or three tabs. We will be working primarily in the Variable view.
There are three descriptive outputs we are going to learn how to request: Descriptives, Frequencies and Measures of Central Tendency.
To run these, you will use the analyze tab at the top of the window. Select Analyze>Descriptive Statistics>Descriptives. Once the small box pops-up, you will see a list of the variables on the left and an empty box on the right. We are going to select the following variables on the left to practice with: age category, marital status, level of education, employed, and gender. Once you move those to the box on the left, select the options box on the right and make sure the following are selected: mean, minimum, maximum, and std. deviation. Then, click continue, and then click ok.
A new window will open, this is called the output window. It should display a small table that looks like this:
This tells us a few key pieces of information. First, the N represents the sample size or the total number of cases in our sample. Minimum represents the lowest value for the given variable. You can see that some are 1 and some are 0. Those are the lowest for the given variable. The maximum represents the highest value for the given variable. The mean is the average of all the cases ranging from the minimum to the maximum. The Std. Deviation tells us how much one score deviates from another, or how far apart the largest deviation is. A higher std. deviation indicates a large spread or that the cases are spread out from another. A smaller std. deviation indicates a smaller spread or that they are centered around the mean. This information is helpful in understanding the data and the individual variables we are working with. The variables we selected are variables we would consider demographic variables or variables related to the identity of individuals. So, what we can learn from this descriptive output would be reported like this:
In a sample of 1000 individuals, the mean gender for the sample was 0.16 on a scale of 0-1. The mean employment status is 0.61 on a scale of 0-1 with an average education being 1.8 on a scale of 1 to 4 with a standard deviation of 0.8. The average marital status is 0.42 on a scale of 1 to 4. The average age category selected was 2.23 on a scale of 1 to 4 with a standard deviation of 0.956.
So, not too descriptive but we know something. If we knew the categories for each we would be able to report more information. Let’s look at frequencies to see if we can gather more information about our sample. Go to Analyze>Descriptive Statistics>Frequencies.
A pop-up will appear again. Select the same variables we used prior and move them to the variable box on the right. Then select the box that says Statistics. Select the following: mean, median, mode, std. deviation, minimum, and maximum. Click continue, then click ok. You will now see a series of tables. The first table should look like this:
This table contains all the variables and the measures of central tendency and dispersion we requested. It is very similar to the descriptives but is reported differently. The remaining tables provide us with more information. Let’s take a look at them. The remaining frequency tables report the categories and the counts or frequencies for each. It also includes the measures of central tendency (mean, median, and mode) and the measure of dispersion (standard deviation) we requested. We now have more information to report. Let’s look at the first frequency table:
We can now better understand the age category because we have the category descriptors. Using the information in the Descriptives output previously, we can see that the minimum value for this variable is 1 and the maximum is 4. We can also see that our mean is 2.23 with a standard deviation of 0.956. Our write-up will look differently now that we have more information. Here is what you would report in your write-up for this variable:
In a sample of 1000 individuals, the minimum age category was “18-21 year olds” with 256 individuals identifying in this range or 25.6% of the sample. The most frequently selected age category was “22-29 year olds” with 374 individuals identifying in that range or 37.4% of the sample. The third age category had 257 individuals or 25.7% of the sample indicating between “30-39 years old”. The final age category had 113 individuals indicating 40 or older. This comprised just 11.3% of the sample. The average age category selected was 2.23 with 63% of the sample indicating they were between 18 and 29 years old with the most frequent selection being between 22-29 years old.
This is just one way to report the information, there are other variations that are acceptable. The important point is to convey the information in a descriptive way so that we can get an idea of the data. From this, we can learn that 63% of the sample are between the ages of 18 and 29. We also know that just 11.3% are 40 or older. What this tells us is that we can’t make predictions about people that are younger than 18 and we can’t generalize about those over 40. But we can make predictions and generalize about people between 18 and 39. Practice making interpretations with the remaining tables.
To complete this step for the exercise, use the data you downloaded to run descriptives and frequencies using demographic data. Identify the demographic variables from your data and choose at least five (5) to run the descriptives and frequencies. Once you run them, draft a write-up to report the data for each. Include at least 3 tables or charts with your write-up. These tables and charts should be ones you create using excel or another software. Be sure to format them using APA citation style.
