Data cleaning

You collected your quantitative data, now comes the fun part: finding out how your study came out! I will start the discussion of analytic strategies with the basics of quantitative data cleaning and analysis. Start by thinking about how data gets into your SPSS or other statistical software program, there are several options. The first option is to use Survey Monkey or other surveying software sites. They allow you to directly download your data into an SPSS file. You will need to double check everything is as you expect, but in general, the data do tend to be accurate.

The second option is the old-fashioned way of entering the data by hand. This is commonly used when you have used paper surveys/ instruments. The concern for this method is it is easy to mistype and introduce errors into your data set. It is a good idea to have someone check your work when entering data. I also recommend doing some descriptive statistics, looking at the range of the scores for each variable to make sure they are as expected (e.g., if you are expecting scores ranging between 1-5 a 55 tells you have entered a wrong number).

What do I mean by data cleaning? There are many definitions, but I am talking about a two-step process. First, double-check all cells are filled, you will probably discover some are not and decisions will need to be made on this. The second step is carefully checking the statistical assumptions of your variables and looking for extreme scores.

Why are these steps necessary? Because the results of your study will only be as accurate as the data you analyze. Therefore, it is very important to take the time to check your data carefully, so you know your results are valid and accurate. I want to refer you to a great book that much of my advice over the next few topics will be based: Osborn's (2013) Best Practices in Data Cleaning. 

Next time we will consider missing data. Do you have an issue or a question that you would like me to discuss in a future post? Would you like to be a guest writer? Send me your ideas! leann.stadtlander@waldenu.edu

Recruiting Interview Subjects

Previously, I discussed you should not recruit people you know for interviews, so how do you recruit people you do not know? There are two common ways, which are typically approved by the IRB:

1) Work through an organization. You can work through an organization to access their employees/ members/ customers/ or patients. Typically, the IRB does not want you to contact personally potential interviewees, instead having the organization forward an email with a description of your study information often works. Alternatively, posting flyers or having handouts available onsite is an option. You will need the organization's permission through a letter of agreement.

2) You may be able to use a snowball recruiting technique in which you give out flyers/ send out a description by email of your study to people you know. You then ask them to forward on the information to people they know. Remember, you should not know your final interviewees.

Next time we will consider cleaning your quantitative data. Do you have an issue or a question that you would like me to discuss in a future post? Would you like to be a guest writer? Send me your ideas! leann.stadtlander@waldenu.edu

The Problem with Easy Interview Subjects

As a beginning interviewer, you may want to find the easiest way to recruit potential subjects. Students often want to select people with whom they already have relationship, such as: family, friends, coworkers, students they know, or others they may know. While it is understandable to want to use people you know, it is problematic and complicates, and even contaminates, your interviews.

Family and Friends

Sometimes it is tempting to use family or friends as participants in the study; however, this carries with it some serious issues. First, your prior relationship influences the comfort level they have with you, as the interviewer. Thus, they may be more likely to reveal things to you other people would not, providing a false picture of your sample. Similarly, they may be uncomfortable answering personal questions others would answer (picture asking your grandmother about her sex life, not very comfortable for you or her!).

The other issue with friends and family is you may assume you know what they are talking about and not explore in the depth needed in a research interview. It is a good rule that you do not personally know your subjects.

People You Supervise

Conflicts of interest occur when you interview people you supervise. An example might be a principal interviewing the teachers she/he supervises. However, in the situation of interviewing supervisees, you are placing them in a difficult position. They cannot risk be totally honest with you, if you are also in a position to affect their job. It is advisable to seek similar individuals you do not supervise.

As you can see, there is no easy way to recruit participants. Do not fall for the trap of thinking you can "cheat the system," you will only waste your and your participants' time in conducting unethical research. 

Next time we will consider how best to recruit subjects. Do you have an issue or a question that you would like me to discuss in a future post? Would you like to be a guest writer? Send me your ideas! leann.stadtlander@waldenu.edu

Bias in Qualitative Studies

Jane asks: "how do you avoid bias in qualitative interviewing and analysis of data?

Great question, Jane! There are some safeguards built into the process, if you follow the standard qualitative protocols (e.g., see Creswell & Plano, 2011). Here are some general guidelines. First, make sure your research questions are neutral, so you are not predicting a particular outcome. For example, "What is the experience of homeless people in dealing with the medical community?" is more neutral and less biased than "Do homeless people believe the medical community is dangerous?"

Second, when you write your interview questions, remember they must follow from your research questions, be open-ended, and they must be neutral. You should not lead the participant into a particular belief you have about them, instead you want the participant to say or indicate whether the issue is important. An example would be, "tell me about your last medical visit." This is much more neutral and more likely to get you their opinions than "do you think all doctors are ageist?"

The third way to reduce your bias is to keep a journal (or field notes) where you write down your opinions and insights. Here is where you get a chance to note your biases and internal beliefs. Do not express them to your participants.

Fourth, when analyzing data it is always a good idea to have a second independent person go through and code your data, or at least double-check your coding of responses. This is not a job for a spouse or close family member, instead consider a dissertation peer. If you do not know anyone, check if your committee can suggest a student. You want the independent coder to not know how you are expecting the data to come out, keep them neutral so they can spot any biases you have introduced. 

Next time we will consider "easy" interview subjects. Do you have an issue or a question that you would like me to discuss in a future post? Would you like to be a guest writer? Send me your ideas! leann.stadtlander@waldenu.edu

A Three Interview Series

Phenomenological interviewing requires an understanding of both the context and the meaning a participant associates with a particular phenomenon. Seidman (2013) recommends the use of a three-interview series, in order to delve deeply in the context and to establish trust with the participant. Interview 1 establishes the context of the participant's experience. Interview 2 allows the participant to reconstruct the details of their experience within the context in which it occurred. Interview 3 encourages the participant to reflect on the meaning of the experience. While, interviewing your subjects three times may not be possible, you can still approach a single interview by bringing in characteristics of the three-interview series detailed here.

Interview One: Focused Life History. In the first interview, the interviewer's task is to put the participant's experience into context by talking about him or herself in light of the topic of interest. The interviewer avoids the use of "why" questions, using "how" questions instead. For example, "how did you come to be in an online dissertation program?" By asking "how" you are leading the person to reconstruct the event and place it into the context of their family, school, and work experience.

Interview Two: The Detail of Experience. The second interview concentrates on the details of the participant's current life in the topic area. Here you are collecting details upon which their later opinions will be built. As an example, you might ask online dissertation students what they do each day related to their dissertation. You are trying to get the person to reconstruct the many details of his or her life that make up the experience of interest. Thus, you might ask about the dissertation student's relationships with his or her chair, committee member, other students, spouse or partner, children, work colleagues, and friends. You might ask them to reconstruct a day in their dissertation life from the moment they woke up until they fell asleep. You may ask for stories about their experiences in school as a way to of eliciting details.

Interview Three: Reflection on Meaning. In the third interview, participants are asked to reflect on the meaning of their experience. It addresses the intellectual and emotional connections between the participant's work and life. An example question with an online dissertation student might be: "Given what you have said about life before you started your dissertation and what you said about your dissertation work now, what does completing a dissertation mean to you in your life?" Making sense or meaning requires that the participant look at how the factors in his or her life interacted to bring him or her to the present situation. 

Next time we will examine bias in qualitative studies. Do you have an issue or a question that you would like me to discuss in a future post? Would you like to be a guest writer? Send me your ideas! leann.stadtlander@waldenu.edu

Phenomenological Interviewing: Philosophy

Phenomenology is actually a philosophical approach to interviewing. Different researchers have slightly different methods and views for what they call phenomenology, I will be talking about the view set forth by Seidman (2013). Phenomenology is examining the experience of participants and the meaning they make of it; this view stresses the transitory nature of human experience. As Seidman states (p. 16) "in human experience, the 'will be' becomes the 'is' and then the 'was' in an instant." Therefore as interviewers, we are interested in the participant's "is," recognizing that it may change in the future.

Consider in most of life we are seeing the world from our point of view, which may be quite different from how others view it. In phenomenological interviewing, the goal, as the researcher, is to set our own beliefs about the world aside and understand the world from the interviewee's viewpoint. We will never fully understand their view, but we must strive to come as close as we can.

Researchers emphasize that a phenomenological approach focuses on the "lived experience" of others. What does "lived experience" mean? It means both the interviewer and interviewee step back and reflect on the reconstruction of what the interviewee felt during the phenomenon of interest. Interviewers strive to guide the interviewee to recreate mentally their lived experience, trying to make the "was" into the "is."

Finally, a phenomenological approach emphasizes the meaning of experiences. A basic assumption is that the meaning people make of their experience affects how they live the experience. By asking participants to reconstruct their experience and reflect on it, researchers are asking people to give meaning to it. It is the context of the experience: their feelings, rationale, and thoughts about the experience, which gives it meaning. Therefore, understanding the person's words that he or she chooses to use become very important; clarity of meaning is the goal. As an interviewer, you must put the experience into the wider context of the person's life. For example, knowing a woman is currently a dissertation student only can be understood when you understand what it means to her in the context of her life.

Next time we will consider doing a 3 interview series. Do you have an issue or a question that you would like me to discuss in a future post? Would you like to be a guest writer? Send me your ideas! leann.stadtlander@waldenu.edu

Nominate your Mentor!

Walden University is now accepting nominations for the Summer 2017 Faculty Excellence Award.

The Faculty Excellence Award is presented up to two times annually at the college level, recognizing outstanding Walden University faculty members who exemplify the university’s core values of quality, integrity, and student-centeredness.

Eligibility and Criteria

Eligible nominees for the award must be current core or contributing faculty members, or program coordinators with teaching as a primary responsibility, with a minimum of two years of service to Walden University. Program directors, associate deans, and staff are not eligible. Nominees must have demonstrated the following: 

• Mentoring and teaching excellence
• Service to the university
• Service to the community and to organizations
• Scholarly contribution to his or her discipline

Previous recipients of the Faculty Excellence Award, Extraordinary Faculty Award, and the Presidential Award for Faculty Excellence are not eligible for this award. Visit the Faculty Excellence Award webpage for a list of past award recipients.

Nominations

Nominations will be accepted from Walden students, alumni, and faculty members. To nominate a faculty member, complete the FEA Nomination Form by March 31, 2017. Nominations received after March 31, 2017 or incomplete nominations, will not be considered.