Friday, July 30, 2010

Research Proposals and Confusion about Dependent Variables

The following blog posting was originally uploaded to my hub here: http://hubpages.com/hub/Research-Proposals-and-Confusion-about-Dependent-Variables

Having taught a year-long, two-course research component to Kean students for the past few years, it has become apparent that one of the most nagging issues for students is the dependent variable. Consider the following:

"My research proposal is about whether the death penalty reduces murder, so my dependent variable would be whether or not states have capital punishment, right?"

Wrong.

Research methods textbooks aren't necessarily helpful. Yes, they give the definition but even when my students do the reading, I can see from the blank stares that what they've read isn't sinking in. Here's a sampling of definitions for dependent variables from a few methods books:

From Singleton & Straits (2010), Approaches to Social Research: "The dependent variable is the one the researcher is interested in explaining and predicting. Variation in the dependent variable is thought to depend on or to be influenced by certain other variables. The explanatory variables that do the influencing and explaining are called independent. If we think in terms of cause and effect, the independent variable is the presumed cause and the dependant variable is the presumed effect."

From Maxfield & Babbie (2011), Research Methods for Criminal Justice and Criminology: "The variable assumed to depend on or be caused by another variable (called the independent variable). If you find that sentence length is partly a function of the number of prior arrests, then sentence length is being treated as a dependent variable."

And from my beloved Bachman & Paternoster (1997) Statistical Methods for Criminology and Criminal Justice: "Dependent variable - the variable that is being affected or influenced by another variable. It is often denoted as y. In a causal analysis the dependent variable is caused by the independent variable."

See a common theme in these definitions? The variables pertain to cause and effect. In the social sciences (like CJ, sociology and psychology), the effects are often people's behavior. In criminal justice field, those behaviors of interest would be..... crime! Crime. Antisocial behavior. Violence. Bad things that people do.

How do we measure our dependent variable crime? There are various ways, including arrests, victimization data (people reporting that they were harmed), and self-report criminal behavior (what people tell you they did criminally, for which they may or may not have been caught).

So to return to the original research question "Does capital punishment reduce crime?" the dependent variable would be crime, specifically states' arrest rates.
Other examples:

(1) Research question: Does the such-and-such conflict resolution program make schools safer?

Dependent variable: measure(s) of school safety such as # of fights, # of times the police are called to a school, # of suspensions for fighting.

(2) Research question: Does Colorado's anti-bullying law reduce bullying in school?

Dependent variable: measure(s) of bullying in school, such as data gathered through surveys of students about whether they've been bullied, done any bullying, or witnessed any bullying

(3) Research question: Does mandatory arrest by police reduce domestic violence (DV)?

Dependent variable: measure(s) of domestic violence reoffending (for instance)

(4) Research question: Which makes neighborhoods safer, having the police drive through the area or walk the beat on foot?

Dependent variable: measure(s) of neighborhood crime, such as home burglaries

Wednesday, July 14, 2010

Welcome Back! Looking ahead to Fall 2010 Seminar

To my many, many incoming CJ_4600/Senior Seminar students:

Welcome to the course! I look forward to meeting all of you in just over a month. Some of you I already know, either from Research Methods or another course. In any case, welcome, welcome, welcome.

Two words: don't worry.

Senior Seminar is the course many students dread. Why? It's a lot of work. You probably haven't done anything like this - execute a research study - before. You've heard horror stories from other students.

I'll try and reassure you - it'll be fine. Really. You'll live through it. There will be a lot of you, but I try to give each student about 30 minutes of one-on-one time throughout the semester.

Let me give you a quick picture of the shape of the semester. The first four weeks are spent gearing up to submit your research proposals. If you haven't done so already, you'll take Kean IRB's human subjects protection course and submit to me your course completion certificate. You'll also do a series of in-class and homework exercises that will give you a feel for the various research methods you'll have the option of using for your study. By about week five of the course you'll submit to me your research proposal.

Very Important Point: While this proposal will undoubtely be similar to the proposal you submitted in CJ_3675, it will not be identical to that proposal. CJ_4600 is a different course. You'll need to revise your CJ_3675 proposal to fit the requirements of CJ_4600. We'll go over this more in class. This is just a heads-up.

I typically grade the proposals in about a week, then return them to the students. You'll either be Approved to Begin, Approved-With-Revisions to Begin, or Not Approved. For the non-approved students, you'll have a week to fix the problems and then you'll meet with me again to explain what you did. By about the 7th or 8th week in the semester, everyone should have begun their data gathering.

By about the 10th or 11th week in the semester, the entire class goes to the computer lab to begin data entry and preliminary analyses. In my experience, this is when most students need a lot of hand-holding from the professor. I will get to each and every one of you, and only ask that you remember that there's only one of me and 25 of you in every class. So just be patient. I'll get to you, I promise.

By week 13 or 14, the final papers are due. During the last two classes, students present their findings to their fellow classmates and me. I'll send everyone a template powerpoint presentation to use as a model for their own slides.

One final point, which you'll probably hear me say in class too - everything I'm asking you to do, I've done myself. Proposal writing - check. Surveys and focus groups - check. Content analysis - check. Analyzing existing databases - check. This summer I'm doing both a content analysis and analyzing data from an existing database. I wouldn't ask you to do anything that I myself didn't have a hands-on feel for, if that's any consolation.

In a nutshell, that's it. I promise you'll survive and move on to graduation. And if you have any plans to attend graduate school (hello, Kean MA in CJ program http://www.kean.edu/~keangrad/CHSS/MA_in_Criminal_Justice.html), having done an original research study like what you'll do in CJ_4600 will help your application. If you like doing the kinds of research work we'll do in Seminar, then graduate school might be for you.

Enjoy the rest of your summer, and I'll see you in September.

Best,

Dr. Hassett-Walker

Tuesday, March 2, 2010

Content Analysis: Next Steps

You're proposing to do a content analysis. Dr. Hassett-Walker has approved (or approved with revisions) your study. Now what?

Here's what you do next (i.e., beginning your study):

(1) Do the search of your selected content. Ex: Searching in Westlaw for past articles in the NY Times and the Washington Post, during the years 2000-2010. Your search terms are "university" and "campus" and "violence" and "victims" (for a study - my hypothetical example - about the effect of violence on college and university campuses on primary and secondary victims).

Note: your search terms will be similar, probably, to some of your code list words.

Suppose 8,547 articles come up. You don't need that many. You need 100, but will randomly select 200 to account for some articles (let's say every other one) being non-relevant or less relevant.

(2) At Randomizer.org, generate a random number list. I just did this, and told Randomizer to give me 1 set of 200 randomly generated numbers, ranging from 1 through 8,547. Here's part of the list it gave me: 49, 2387, 4163, 3955, 4424, 2349, 77, 2422, 3480, 128, 5972, 1579, 3784, 2312, 624, 5602, 876, 2977, 2948, 6856, 2736, 7490, 2117, 7590, 4569, 2844, 2071, 5033, 6005, 2112, 4073, 3580, 7596, 7362, 7066, 1793, 5914, 2815, 7584, 3384, 118...

(3) Next, I find those articles that correspond to the numbers in the random number list. I copy/paste them into a Word or Word Perfect document.

(4) Now I begin reading the articles. Working in Word (rather than on paper), I use the Word artist function to color-code red all words and phrasings related to violence and shootings (e.g., shooting, gun, shots fired, etc.). I color-code green all words and phrasings related to victims (e.g., victim, victims, witness, students standing there watching, etc.) I color-code yellow all words and phrasings related to college or university campus (e.g., campus, college, university, classrooms, buildings, etc.). I color-code blue all words and phrasings related to programs the university or college put into place to help victims.

(5) I set up a database skeleton in SPSS. Before I can enter any data into the database, I must create my variables. These will correspond to my code list. EXAMPLE of variable name (variable #1) and corresponding label: ArticleNum, which newspaper did it comes from (1=NYT, 2=Washington Post). Next variable name: VictimNum, number of times victim or victims mentioned in article. Next variable name: VictimLang, language used to describe victims. Etc.

Note: any variables that involve counting will be numeric variables. Any variables that will have me inputting words will be string variables.

Remember: You set up your SPSS database in variable view.

(6) Once I have my SPSS skeleton set up, I can enter my data (words, language, counts of words) into SPSS. I can either do this once I'm done reading all the articles, or as I go. Remember: you enter your data into your database in data view.

(7) Once you've entered all your data, you run your preliminary and "main" analyses: frequencies, descriptives, and perhaps a correlation analysis. Review your results with Dr. Hassett-Walker. Do you think you are finding support for your research question or hypothesis?

(8) Once you have determined which relationships are significantly related to each other, go back to the color-coded newspaper articles that make up your data. Search for the text coded with the code words that pertain to those statistically significant relationships. That's the text you want to read and perhaps copy/paste into a separate word document. Boil it down into a succinct one to two-paragraph summary/summaries that further explains the statistically significant relationships you found with the quantitative data.

Survey and Focus Group Methodology, Analyses, Results

For the students doing the survey and focus group option, once I've approved (or approved with revisions) your study, you're ready to start.

(1) Print out your consent form on Kean letterhead, which you'll get from me. You and I both sign in the appropriate places. Then you make around 50 copies (two for each student/human subject, one of which they sign and return to you, the other of which they keep).

(2) Make 25+ copies of your survey instrument as well. Your subjects will be filling this out, so it has to be in a form that's ready-to-use for them.

(3) Next step: subject recruitment. You should have already been thinking about this, and/or written it down as part of your method section. Where and when will you find your fellow Kean (NO OCC students!) students? Student Center? Library? Around campus? How will you approach them? If they say yes, you should hand them a consent form to sign. Once they've read, signed, and returned the consent form (keeping a copy for themselves), you'll need to take them to the location where you'll conduct the focus group. Plan and pick your location ahead of time. It should be somewhere you're allowed to be (empty classroom? Empty study room in the library?) that is quiet and has access to outlets, if your tape recorder needs to be plugged in.

(4) Equipment: Also plan this ahead of time. You need tapes that run long enough (1 hour should be good), maybe extention cord, extra batteries, and perhaps an extra tape recorder.

(5) Visualize where you will put the tape recorder, and how your subjects will sit around it. They should sit in a circle, with the tape recorder in the middle.

(6) During the focus group, periodically check that the tape recorder(s) are running, that they haven't timed out.

(7) Immediately after each focus group, spend some time writing down everything you can think about that happened during the focus group. What was the mood? Was everyone equally talkative? Were some students (probably) more talkative than others? What were some of the key points discussed? Write as if you might discover that neither of your tape recordings worked, so you want to be as thorough as possible in writing down what went on and what was discussed.

(8) Once you have all your surveys collected and focus groups conducted, you enter the survey data into a SPSS database that you create. (Email Dr. Hassett-Walker if you need help setting up a skeleton for this.) You'll also spend a few hours typing up everything that's on the tapes. (Play-stop-rewind, play-stop-rewind, play-stop-rewind.... Yes, it's a tedious process.) In the end, you should have a transcript that's perhaps 60-75 pages long. If you ever took a drama class in high school, what you produce will be similar to a script for a play. Note: You don't have to capture every "um" and "ah."

(9) As a reminder, the Krueger book on conducting and analyzing focus groups is on reserve at the OCC library.

(10) Once all your survey data are entered into SPSS and your focus group transcripts are typed up, you are ready to begin analyzing your data and summarizing your findings. You are doing quantitative (of the survey data) and qualitative (of the focus group transcripts) analyses.

(11) The quantitative analyses: this will be the focus of our lab visits in late March. That said, you already know what these are, as you've run them in Methods and Seminar: frequencies, descriptives, correlations and crosstabs with a chi square test, as appropriate. Also refer to p.11 in your course syllabus.

(12) The qualitative analysis: This will basically involve reading the transcript, and determining what group #1 said about question #1, what group #2 said about question #1, what group #3 said about question #1, etc. If it's easier, you can open up a separate Word (or Word Perfect) document and copy/paste text from the original transcript into that new document, under the subheading of (for instance), "All groups' feedback on question #1." So then you have all your text about question #1 in one place; then all your text about question #2 in one place; etc. You want to read through it, summarize it, shorten it so that you have a short, concise 1-2 (or longer) paragraph summary of the key points and themes from the focus group discussions, and a few quotes that "pop", that describe what your subjects felt about the issue that was question #1 (and question #2, and question #3, etc.). Ask yourself how the focus group discussions (your qualitative data) reinforce, condradict, or further clarify your survey findings.

In your final paper, each summary paragraph of the qualitative data will go after each relevant quantitative data table (e.g., of frequencies or correlations), and your paragraph summary of what's in the table.

(13) Example of part of a results section from a paper I wrote:

RESULTS
Fourteen (n=14) Kean University students took the survey and participated in one of several focus groups. Demographically, 79 percent of participants were women, and 21 percent were men. Half the sample (50%) was Caucasian, 14 percent was Black, and 14 percent was Asian. In terms of ethnicity, 21 percent self-identified as Hispanic. Mean age of the sample was 28 years. Socio-economically, nearly 29 percent of the sample self-identified as working class. Thirty-five percent of survey/focus group participants indicated they were middle class, and 35 percent self-identified as upper middle class.

"The guys basically went to an ivy league school”: Class Difference between the Women Performers and the Lacrosse Team Members

Most participants – just over 71 percent – agreed or strongly agreed that the women performers and male lacrosse team members were in different social class groups, as seen in Table 1, below.

Table 1: Women Performers and Lacrosse Team Members were in Different Social Class Groups (n=14)
Agreed or strongly agreed: n=10, 71.4%
Neither agreed nor disagreed: n=2, 14.3%
Disagreed or strongly disagreed: n=2, 14.3%
TOTAL: n=14, 100.0%

[NOTE: THE ABOVE TABLE ISN'T COMING ACROSS WELL IN THE BLOG. IN THE ACTUAL PAPER, IT'S A TABLE IN WORD.]

[HERE’S WHERE MY SUMMARY OF THE FOCUS GROUP FINDINGS COMES IN, INCLUDING SOME SPECIFIC QUOTES]
Some participants felt the women and men were in different class groups because the men “could hire expensive lawyers” whereas one of the dancers, Crystal Mangum, came from a more modest background, being the “daughter of a retired auto-mechanic… that, like, pretty much put her in a family of the working class.” In addition, the fact of the men attending Duke University (“ivy of the South”; and “the reputation of the school… ranked the top ten institutions in the country as far as status, and it’s not cheap”) meant that “these kids have money obviously.”

By contrast, the women worked in a profession (exotic dancing) with low occupational prestige. “The fact that they had to dance. I don’t think that would be a woman’s first choice of occupation.” Another participant explained, “dancers…are automatically associated with being strippers or hookers, or being paid for…you are automatically labeled or looked down upon.” Another participant felt that the men “would have disrespected them [the women] anyway because of their job and double disrespected them because of their ethnicity.”

Another participant said that the women likely danced because they had to in order “to pay bills, as opposed to lacrosse team in Duke University. Most likely, they had their parents pay for their education so I would think they’re in different social classes.” Another commented on the fact that the lacrosse team was able to hire the women (“it’s like a lot of money, like $500 right off the bat”), suggesting that the employer-employee relationship indicated a class difference between the men (employer) and the women (employee).

THE IDEA IS TO USE THE DISCUSSION FROM THE FOCUS TO FURTHER EXPLAIN OR CONTRADICT WHAT THE SURVEY FINDINGS (PRESENTED IN TABLE FORMAT) TELL YOU. SURVEYS TELL YOU COUNTS AND PERCENTAGES OF RESPONSES, BUT THEY DON’T REALLY GET AT THE “WHY” AND “HOW” AND “WHY DO YOU THINK THAT.” THAT’S WHERE THE FOCUS GROUP RESULTS COME IN.

Friday, February 26, 2010

Downloading Your Existing Dataset

So the professor has approved your study. You're proposing to download an existing dataset from the National Archive of Criminal Justice Data. You're all ready to get started, and... what do you do?

If you did the SPSS download & convert homework, it's basically the same steps. First up, it will be helpful to work in a computer that has SPSS installed. (See point #2 below re: links to free or small-fee license for SPSS, which Sydney [thanks, Sydney!] found.)

(1) Download your dataset using the same steps that were outlined in the SPSS homework. Go through the steps and prompts that it gives you at the NACJD website. Download them into a file with a short path name (the conversion will work better that way, for some reason), for instance c:\Seminar\ and then download your files there. It will likely give you more files than you need (some for SAS software, something with ".por" at the end maybe). You need the setup/syntax file with ".sps" at the end, I believe.

(2) To complete the conversation of the ".txt" file into an SPSS database, you'll need to be working in a computer that has SPSS. You can either get a free or minor-fee version of it downloaded into your home PC or laptop, or save your files from NACJD onto a memory stick, and bring that to the OCC computer lab.

Here are the links to SPSS downloads, either free or for a minimal fee for a six-month license. Note: I haven't tried either of these, so I can't vouch for them. That said, give them a try.
http://www.brothersoft.com/spss-download-72110.html
http://www.onthehub.com/

(3) Open SPSS. Then go into File-Open, and you want to click the drop-down arrow so you can Open Syntax. If the correct directory doesn't come up, find your C:\Seminar\ directory (or whatever you've called it), and the SPSS syntax file should be there. Open that up.

(4) The next step is the replace the words "physical-filename" (in DATA LIST FILE="physical-filename" / ) with the full path name location and name of the data file. NOTE: Make sure you're in the top (not the bottom) of the syntax file. For some reason, the same language (with "physical-filename") is also located at the bottom of the syntax file. That's not the one you want. You want the one at the top of the syntax file.

(5) Minimize SPSS and open up MyComputer. Find your C:\Seminar\ folder (or whatever you've called it). Now you want to right-click (probably) over the ".txt" file that is the pre-conversion SPSS file, and from Properties get the location ("C:\Seminar\ ") and the file name (something like "08381-0002-Data.txt") of the ".txt" file. Next, back in the SPSS syntax file, the location and file name get copy/pasted in between the quotes to replace "physical-filename." You may have to do this in two steps (first to copy/paste over "C:\Seminar\" and next to copy/paste over the name of the .txt file.) I know it sounds confusing when I write it out like this, but if you were able to successfully complete the SPSS homework, this is basically the same process. To complete this step, you'll be going back and forth between MyComputer and the SPSS syntax file.

(6) Once you've got the location and file name pasted into the syntax, click the "run all" button. An SPSS output window should pop up. In the lower right corner of either the output window or the SPSS database window, you should see the words "transformations pending." Click on "Tranformations" or "Tranform" and then "Run all" (something like that).

If you DON'T see "transformations pending" in the SPSS or output window (some people do, some people don't), then go back into the syntax file and click Transform-Run Pending Transformations (something like that). In either case, the next step is to run the pending transformations, which should upload the data into SPSS. (Before you complete the transformations, all you have are the variables in a blank SPSS database skeleton. You need the data, too.) This process gets completed in either the SPSS or output window, or the syntax window.

(7) The SPSS data should now be ready to work in.

(8) Save the database under a new name. Remember the location of your database. (On your memory stick? On your home computer or laptop? Don't save it on the OCC lab computers.) Remember that for when the class goes to the computer lab in late March, you'll need to transport your data on memory stick, so we can work on it together there. Everyone gets one-on-one data analysis help with Dr. Hassett-Walker in March.

(9) Next, start locating and renaming your variables. The variables will probably have unhelpful acronyms like ACB8xy2, which really means "Gender." So rename it as "gender." I also find it helpful to fill in the labels for my chosen variables, if SPSS didn't do that for you in the conversion.

(10) Your database may have 892 variables (or whatever in it). You want to select up to 15 (and fewer, eight or ten or twelve, is fine). Locate your chosen variables, and run some frequencies. Look at the results. Any missing data? Any data coded for "did not respond"? Things like that will need to be filtered out. We may also want to cut out some of the variables that you're not using (i.e., make the database smaller), which will make it run faster.

(11) When you've gotten through all this, email me (if you want to at that point email me your dataset and frequencies, that's fine) and I can talk you through some next steps. We'll also do the next steps at the lab in late March.

Friday, January 15, 2010

K2/evening Seminar students, blog assignment #2 (due March 24th)

K2/evening Senior Seminar students - it's that time again. Time to post your second blog-comment status update about the progress of your research. (Try to contain your excitement.)

You've all been working very hard, and the end is in sight (April 7th). Hurrah! You should be done, or close to done, with your data collection and have begun to enter and/or analyze your data. Then comes the fun part! (Seriously.) Finding out whether or not your hypothesis is true, finding out the answer to your research question.

I look forward to reading your comments. As you did before, post them in the "comments" section of the blog.

K2/Evening Senior Seminar Students, blog assignment #1 (due Feb 10th)

Students enrolled in my K2/evening section of Senior Seminar (which meets Wednesdays at 4:30pm) should post here a status update re: where they're at with their proposal. Proposals are due next week, so at this point you (the student) should have a solid idea of what-all your study will entail.

Post an update - correct spelling and grammar, please - in the "comments" section. I look forward to reading your updates.

If someone raises an issue that I think is applicable to everyone, I'll post a reply for all to read.

Good luck!

K1 Senior Seminar Students, blogging assignment #2 (March 24th)

It's that time again. Time to post your second blog-comment status update about the progress of your research. (Try to contain your excitement.)

You've all been working very hard, and the end is in sight. Hurrah! You should be done, or close to done, with your data collection and have begun to enter and/or analyze your data. Then comes the fun part! (Seriously.) Finding out whether or not your hypothesis is true, finding out the answer to your research question.

I look forward to reading your comments. As you did before, post them in the "comments" section of the blog.

K1 Senior Seminar students, blogging assignment #1 (Feb 17th)

Students enrolled in my K1 section of Senior Seminar (which meets Wednesdays at 11:00am) should post here a status update re: where they're at with their proposal. Proposals are due next week, so at this point you (the student) should have a solid idea of what-all your study will entail.

Post an update - correct spelling and grammar, please - in the "comments" section. I look forward to reading your updates.

If someone raises an issue that I think is applicable to everyone, I'll post a reply for all to read.

Good luck!