When running repeated measures data, it is necessary to organize the data so that each row represents an individual case. Sometimes in the raw data, the rows will organize as individual measurements for the same cases. The example below represents two participants with two outcome measurements (weight and calories), measured at three different time points. There are two formats for representing repeated measures data: the long and wide format. The long format utilizes multiple rows for each observation:
The wide format utilizes one row for each observation or participant:
Using SPSS, the data can restructure from long format into wide format. Here we present the steps for doing this using the above example data.
1) From the Data menu, select Restructure…
2) Then select “Restructure selected cases into variables”.
3) For the “Identifier Variable(s)” box, transfer over the ID variable. For “Index Variable(s)”, enter the designator for the repeated measurement (in this case, it is TIME). Then click Next.
4) Select “Yes” to “Sort the current data”.
5) Click Finish and the data should be restructured to a wide format. Save the file as a transposed version. Note how the variables are automatically given a different name based on the number of measurements. With the data in this format, you can now properly conduct repeated-measures analyses (such as dependent samples t-tests or repeated-measures ANOVAs).
Qualitative research is an iterative process. However, the methods that we use to establish trustworthiness are often separated from the research process itself, leading to post-hoc evaluations of trustworthiness. Instead, these methods should be built into this iterative research process along with other design aspects like instrumentation, methods of data collection, and analysis. The good thing is that this is easy to plan and do, as it simply involves the already active, dynamic relationship that you have with your research.
One way to ensure trustworthiness of a study is by continually reviewing your research methods against your research questions. As you are move through data collection, are your methods yielding the information that you need to address your research questions? Will these provide you the data you need for your analysis? If you are conducting interviews, reviewing the data after each interview would be appropriate.
Another way to establish trustworthiness as you are collecting data is to ensure a sufficient sample size. You may already have a target sample size; however, it is helpful to remember that for many studies (though not all) ensuring saturation of the data is important. Again, reviewing collected data after each interview can help you determine if you have reached saturation or if you need to keep collecting data.
It also may be appropriate to conduct some analysis of the data while in your data collection phase. This strategy often is associated with grounded theory studies. However, analyzing data as they come in not only helps with ensuring adequate sample size and saturation, but provides a continued check that the method is appropriate for the research question, that the method is appropriate for the analysis, and that all of this is congruent with your research questions.
Remember, your research questions drive your method and analysis, and these can be flexible. Qualitative research is a dynamic process with a lot of back and forth and a lot of reviewing, thinking, reflecting, and tinkering. And this is okay! Recognizing this and working with it, rather than forcing your study into a box based on your proposal (especially if things are not working as initially planned) will lend itself to a solid study that has trustworthiness built in.
In past blogs, we have discussed interpretation of binary logistic regressions, multinomial logistic regressions, and the more commonly used linear regressions. In this blog, we will discuss how to interpret the last common type of regression: ordinal logistic regression. It is important to note that, although there are other forms of regression out there, most of these are interpreted in the same way as the aforementioned types.
Although ordinal logistic regression involves some of the same steps of interpretation as the other methods, the interpretation of the individual predictors for ordinal regression can be a little tricky. If you haven’t read up on the other common regression interpretations, make sure to review them to stay caught up!
Like previous regressions, first check the model fitting information to ensure the overall regression is significant. You can also investigate the Nagelkerke pseudo R2, which you interpret similarly to other R2 statistics. Ordinal logistic regression differs in interpreting individual predictors. It shows how much each predictor moves the outcome closer to the next “jump up” or category increase.
The way you do this is in two steps. First, identify your thresholds’ estimates. You will have one for each possible increase in the outcome variable. For example, with low, medium, and high categories, you have two thresholds: low to medium and medium to high. Take note of these threshold estimates. You will be using them in comparison to the estimates for each significant predictor variable. For the significant variables, the variable’s estimate represents how much closer they get to a threshold.
Let’s look at an example where we assess students for College Readiness (an ordinal dependent variable), with MATH score and READING score as predictors. The dependent variable ranges from low, to medium, to high readiness. The threshold estimate assigned to low is 5, and the threshold assigned to medium is 10. Once a student hits a threshold of 5, they move to the medium group. Once they hit 10, they move to the high group. MATH score is the only significant predictor with an estimate of 2. This means each 1-point increase in MATH score moves students 2 points closer to the threshold.
So, a student with a math score of 3 is expected to be in the medium group because they tend to move 2 units closer to the threshold for each additional point in MATH (2 units closer to threshold for each MATH point * 3 MATH points = 6). Because 6 is greater than the threshold of 5, that student has broken into the medium category. If their MATH score were 3 units higher (i.e., 6), we would see the following happen: (2 units closer to threshold for each MATH point * 6 MATH points = 12). This would push them past the threshold of 10 into the high group.
You can interpret each significant predictor this way, and even consider how close they get to each threshold based on a combination of predictors, so if READING were also significant, the addition of their score in reading might also help push them past the next threshold even if their math score just barely missed pushing them past the jump.
When it comes to the theoretical foundation of your study, my suggestion is to keep it simple. Sometimes people want to use two, three, or even four theories. That means you will not only need to define each theory but also relate each one to your study and discuss each of them in relation to your findings in the Discussion chapter. All of this can be challenging enough for a beginning researcher using only one theory. Using multiple theories may not only create unnecessary work for you, but also may be unnecessary to adequately frame your study.
I suggest picking just one well-suited theory. The theory should guide your study and include the variables or constructs of your study. If you intend to use two or three theories, you need to have a very good reason for doing so, and all of your selected theories should include the variables presented in your research questions.
You should not approach theory selection as an afterthought. Your choice of theory is related to your research problem and how you develop your study. Your topic should be based on a social issue or problem. Within the literature on the topic, you then work to identify a gap in understanding the topic or a point of contention in the research. This is the research problem. Theories help researchers explain behavior and phenomena. So, there will likely be theories and associated variables related to whatever topic you are studying. These are the theories you want to consider for your study. Some theories may be more suitable for your topic and research problem than others.
Therefore, it is crucial to understand each theory and its components to determine which one relates best to your study. Keep it simple. Use a single theory unless you have a very good reason to use more than one.
An audit trail is a qualitative strategy to establish the confirmability of a research study’s findings. Confirmability involves establishing that the findings are based on participants’ responses instead the researcher’s own preconceptions and biases. Audit trails are an in-depth approach to illustrating that the findings are based on the participants’ narratives and involve describing how you collected and analyzed the data in a transparent manner. Parts of an audit trail may include examples of the coding process (presented in a table), descriptions of how you worked from individual codes to themes, and rationale for what codes were clustered together to form the basis of a theme. The purpose of doing these things is to clarify to your readers why you made the decisions you did and to show that your analysis follows a logical path dependent on the participants’ narratives.
You won’t need to outline every code or theme in the results chapter but can include this information in an appendix. This is ideal for students who want to show the full process. An audit trail is valuable for critical reflection on your decisions, which is especially important in the recursive process of qualitative data analysis.
Finally, the audit trail also is helpful when presenting how you analyzed the data in either the results chapter or the methodology chapter. If you create a logical path for the data analysis, it will be that much easier to report and explain what you did.
Readability is important while applying for Institutional Review Board (IRB) approval for your dissertation study. In this case participants focus on the study materials, which refer to any letters, documents, forms, or instruction. For example, informed consent forms and invitation letters. Write these materials so that your participants can read and understand the information. Readability is crucial in studies, especially with children or populations with lower reading ability than the average adult participant. Depending on your study population and IRB requirements, you may need to provide a readability report to ensure your materials are suitable for participants. In this blog, we demonstrate how to produce a readability report and how to improve the readability of your materials.
Producing readability statistics:
You can use Microsoft Word to produce readability statistics for any text or word document. To do this, first go “Options” under the Word “File” tab. Next, check the “Show readability statistics” box in the “Proofing section” of the options.
Need assistance with your research?
Schedule a time to speak with an expert using the link below.
After you have done that, all you have to do is run a spelling and grammar check on your document. You can do this by clicking on “Check Document” in the “Review” tab.
Word generates readability statistics if the document contains at least one spelling or grammar issue. If all errors have already been corrected, you can insert a deliberate typo, run the check to fix it the typo, and then you will get your readability statistics. They will appear in a popup box like the one pictured below. You can then capture a screen shot of this report to share with the IRB if they require it.
The most important number from the report is the Flesh-Kincaid Grade Level, which corresponds to the United States grade level required to understand the text. For example, the Flesh-Kincaid Grade Level from the report above is 13.5, which means that the text is roughly at a college freshman or sophomore reading level. For research involving nonvulnerable adult participants, it is often recommended that your consent form and other materials be at an 8th grade reading level. Therefore, you would want your document to have a score of 8.0 or less on Flesh-Kincaid Grade Level. If your participants will be children or other individuals with reduced reading ability, your score will need to be even lower.
To improve readability score:
So, what if your readability grade level is too high and you need to lower it? There are two ways to improve readability score: 1) use shorter words and 2) write shorter sentences. The grade level score is calculated based on the average number of syllables per word and the average number of words per sentence. So, using words with fewer syllables and reducing the length of sentences will improve readability.
An easy way to start is to look for long sentences. And see if you can either shorten them or split them up into multiple sentences. Get rid of unnecessary words (e.g., “research study” can just be “study”). Split them up into compound sentences (e.g., “Your participation in the study is voluntary, and you can withdraw at any time” to “Participation is voluntary. You can withdraw at any time.”). Once you have shortened your sentences, look for “big” words and see if you can replace them with smaller words. For instance, if we changed the previous example from “Participation is voluntary. You can withdraw at any time.” to “The study is voluntary.
You can leave at any time.” the readability grade level of these two sentences improves from 9.7 to 4.0! We were able to greatly reduce the score by getting rid of some of the big words (i.e., “participation” and “withdraw”).
I’ve spoken with students lately that had a committee member that didn’t agree with aspects of the dissertation, or a chair that had different views than the AQR. What is a student to do? My advice is to ask questions, ask for help, and move up the ladder.
Ask questions. For example, you can ask your chair, “How different points of views have resolved in the past?” “If you were me, how would you resolve this impasse?” “Would you both willing to hop on a call together to see what common ground can find?” Find ways to ask questions that promote a solution.
Get support
First, seek help from colleagues or other faculty who may be able to assist, whether through their relationships or experience. I recall a time when I faced difficulties with a person at the financial aid department, Lola, during my time as a student. To be honest, I didn’t like Lola, and it seemed that Lola didn’t care much for me either. Every interaction felt like a hassle. So, I decided to reach out to a friend, Roy, who worked in the housing department.
As I explained my trouble with Lola, Roy responded, “Lola is a pussycat.” I was taken aback and said, “No, no, Roy. You must be mistaken. I’m talking about LOLA from the financial aid department.” However, Roy assured me that, yes, we were indeed referring to the same Lola. Once I mentioned Roy to Lola, our relationship changed for the better. Months later, I remember walking past her door one day and pinky waving to each other, and I thought, “boy have things changed!”
Move up the ladder
What if none of these options work out, then what? I think it’s appropriate to take your concern up the ladder to the chair of the department or dean. I don’t think jumping to the dean is a great idea, but if you’re in the middle of things for say 8 weeks, it’s time to reach out to others.
Need assistance with your research?
Select a time to speak with an expert using the link below
That Lola experience was over 25 years ago, but the lesson that relationships can and do change, that people can become helpful, and that the current circumstance is not necessarily the final circumstance sticks with me. Keep this in mind as you move through the dissertation process and graduate. There’s more help around than you know, if you’re willing to open up to it.
You can schedule a free, no obligation consultation by clicking the link below.
I wish you an expedited graduation, and thoughtful planning. You’re almost done, so plan for support early!
Best,
James Lani, PhD
There’s lots of potential roadblocks to finishing your dissertation. The one that I think is demoralizing is the revisions process. There are several reasons revisions can go on for a bit.
Firstly, let’s face it, this is your first and last dissertation, so you don’t know the process. It takes a while to understand what scholarly writing is all about. Faculty too will want to put you through the same scholarly writing paces as they went through. Know that writing strengths vary. Some students are good at doing lit reviews but not good at methodology. These can be places to reach out for help.
Secondly, some things are not in your control. Unfortunately, not all faculty are prepared to advise. Really, all you can do it follow their advice the best you can. Conflicting committee feedback is also something you can’t control (and is frustrating). I think leaning on your chair to resolve these conflicts is best.
Advice to finish-up:
Certainly, look at a dissertation closely related to yours. Look at the headings and organization. In particular, see if you can locate other students’ dissertations that your committee members have accepted and even ask them for dissertations they’ve approved. This can give you an idea of the style that they prefer.
Don’t ever give up—you’re closer than you think.
I wish you an expediated graduation, and thoughtful planning. You’re almost done, so plan for support early!
Best,
James Lani, PhD
Once you have decided on the key terms to define for your study, you will then of course need to define them in your “Definition of Terms” section. The idea of defining the key terms and major concepts of your study is to provide the reader with precise meanings of how you will be using them. This blog provides a few tips for defining your key terms and includes a few pitfalls to avoid.
For your definitions, concision is key. Be brief and precise. Start your definitions with language such as, self-efficacy is… or self-efficacy refers to… This kind of sentence construction is expected and provides consistency among entries. Then, define the term, not generally, but as you will be using it in your study. If they are precise, definitions can be a sentence or two long. Usually, that is all you need.
Ideally, you should support definitions with citations from recent peer-reviewed sources. This lets the reader know that the term and its definition are not arbitrary or general but specific to and recognized in the literature.
Avoid circular logic and using the term you are defining in the definition. Example: “Student retention refers to retaining students in school.” With such a construction, we end where we started and have no better idea of what retention really means as it pertains to your study. If we do not know what retention means, then using essentially the same word retaining is not going to help.
Avoid using quotations in definitions, which is a good rule of thumb in academic writing generally. Paraphrasing is preferred. In something as short as a definition entry, there is no need to use quotes.
Lastly, do not just say that the term or concept is important without defining it. This is a common mistake. Peoples sometimes say, self-efficacy is used to understand… or student retention is important because… but do not actually define the term. Your definitions are not the place to establish the importance of your concepts, and such sentences can be seen as filler. So, make sure you actually state what the term means while leaving out extraneous information.
One section that is often required in a dissertation is the “Definitions of Terms.” This section outlines the key concepts and factors explored in this study, explaining how these elements will shape the research. The “Definitions of Terms” ensures that your readers will understand the components of your study in the way that you will be presenting them, because often your readers may have their own understanding of the terms, or not be familiar with them at all. In this section, you provide a list of terms and define each of them that you will use throughout the dissertation. Seems simple enough, right? But how do you know which terms to define and which ones to leave out?
Include and define terms that are crucial to your study or frequently used throughout the dissertation but may not be common knowledge. You also want to include terms that have a unique meaning within the scope of your study. You do not need to include terms that most, if not all, of your readers will understand without having definitions provided. For instance, you likely don’t need to define ‘leadership’ in your ‘Definitions of Terms,’ but ‘laissez-faire leadership’ would be a good term to include.
However, if your study is about leadership, then it may be beneficial to the understanding of your readers to define leadership based on how you are using it within your study. Things like success or achievement may need definition as well, if you are using them within your study, as the readers will need to know what measures or markers of success or achievement that you will focus on within your study.
For more information on “Definitions of Terms,” including what information to include within the definitions, check out our other blog: How to Write Your Definitions.