What is the relationship between qualitative research and variables? Are they similar to variables of qualitative researches?

In quantitative research hypotheses and variables are tested and measured numerically using statistical methods after the data were collected. Hypotheses and variables exist and stated prior to the beginning of investigation.

In qualitative research no hypotheses or relationships of variables are tested. Because variables must be defined numerically in hypothesis-testing research, they cannot reflect subjective experience. This leads to hypothesis-generating research using the grounded theory method to study subjective experience directly. The grounded theory method uses a data analysis procedure called theoretical coding to develop hypotheses based on what the research participants say. Because the method involves developing hypotheses after the data are collected, it is called hypothesis-generating research rather than hypothesis-testing research. Grounded theory method uses questioning rather than measuring and generating hypotheses using theoretical coding.

Are variables of quantitative qualitative researches similar?

The quantitative research leads to hypothesis-testing research (hypothesis are tested), whereas the qualitative approach leads to hypothesis-generating research (hypotheses are generated).

I would say "concepts" or "categories" rather than "variables" are the concern in qualitative research - with an emphasis on exploring social relationships/social organization, social processes, phenomena, experiences, meaning for different units of analysis and so on (depending upon the particular question and qualitative approach). This is unique from the quantitative focus on defined variables where measurements are the focus. This pdf of a presentation from Kaiser Permanente Northern and Southern California Nursing Research is very basic intro, though there are much more indepth texts available that explore the distinctons:

http://nursingpathways.kp.org/scal/research/resources/researchseries/docs/Part07_VariableWhatYouAreMeasuring.pdf

This is a very basic, and a nursing disciplie-based, angle...others from different disciplines will have different perspectives. Will be interesting to hear from others...

Hi Sonya - a good response here. I too come from a nursing background - but would like to think that the principles are 'generic' to most disicplines. For me - a further consideration, and perhaps where Amrit might have sensed some 'cross-over', relates to 'concrete' versus 'abstract' variables. While, both can be used in quantitative research (although concrete variables are by far the most common) - qualitative research, if it were to refer to 'variables' would be far more inclined to the abstract form. I agree though - 'concepts' or also 'phenomenon' make it clearer.

The benefit of combining qualitative and quantitative methods are as follows:

1. The Quantitative design strives to control for bias so that facts can be understood in an objective way, the Qualitative approach is striving to understand the perspective of the program stakeholders, looking to first hand experience to provide meaningful data.

2. The accumulation of facts and causes of behavior are addressed by quantitative methodology as the qualitative methodology addresses concerns with the changing and dynamic nature of reality.

3. Quantitative research designs strive to identify and isolate specific variables within the context (seeking correlation, relationships, causality) of the study as the Qualitative design focuses on a holistic view of what is being studied (via documents, case histories, observations and interviews).

4. Quantitative data is collected under controlled conditions in order to rule out the possibility that variables other than the one under study can account for the relationships identified while the Qualitative data are collected within the context of their natural occurrence.

5. Both Quantitative and Qualitative research designs seek reliable and valid results. Data that are consistent or stable as indicated by the researcher's ability to replicate the findings is of major concern in the Quantitative arena while validity of the Qualitative findings are paramount so that data are representative of a true and full picture of constructs under investigation.

is your question about "variables" used in qualitative research OR about the "scope" of something similar to variables in qualitative research?

for example the intention of variables in quantitative research is different from identifying 'categories' or 'phenomena'. quantitative research is more 'confirmatory research' (a variable can be used to prove a hypothesis), while qualitative data is used in an exploratory way, following inductive logic.

I agree with Karibu Ringim that both research designs seek reliable and valid results!

If you're interested in mixed methods, I highly recommend Charles Teddlie and Abbas Tashakkori (2009) Foundation of Mixed Methods Research - Integration Quantitative and Qualitative Approaches in the Social and Behavioral Sciences. Sage Pub

I agree with Sonya and Dean in the use of concepts and categories in qualitative research. However, when such research is completed it is possible to extract variables for more concrete measurements using quantitative research.

Hussin - that is correct - at least in the case of 'mixed methods' resarch where methodological triangulation occurs

this is interesting... i like mixed researches and i am always involved with mixed researches. thanks guys for the insights and comments shared.

Teddlie & Tashakkori point out, that triangulation is a 'magical' word and they criticize that, b/c of its overuse, it might not have a 'meaning at all' anymore. They belief, that MM (mixed methods) needs its own terminology (mainly b/c 'triangulation' was introduced to bridge two different 'research worlds'). For example: validity has 35 different meanings within the qualitative/quantitative traditions

Dear colleagues

I do agree that "concepts" and "categories" are useful in qualitative research but

1) They are also useful in other type of research

2)” Concept” is what you need to develop your problem statement as reference to the theoretical framework you use and “category” is what you build for analyzing your data.

Variable is the elementary item (factor or condition…) that is subject to change and that you can inform (not only measure) to validate your hypothesis (ever you are in an inductive or deductive approach). Of course you can have meta variables that can gather several variables or sub variables that can detail some variables.

Unfortunately the step of defining what variables we have to inform is often forgotten in qualitative research. Even and may be more often by people coming from quantitative Research World. That is why, in the courses I give, I insist on defining Variables and the methods to inform them.

You can have open and free access to my distant formation on this point

Sibelet N, Mutel M (2013) Module 2 selecting the survey method and preparing the semi-structured interview: From hypotheses to variables to investigate. In: Sibelet N, Mutel M, Arragon P, Luye M (eds) Qualitative survey methods applied to natural resource management. Available at: https://enquetes-cirad.iamm.fr/course/view.php?id=3

You can also browse or follow the entire course (for free*) on “Qualitative survey methods” you will see it concerns other themes than my own (natural resource management). The course will help you to articulate “concept”, “variable” and “category of analysis”.

Regards

Nicole

*Yes it is absolutely free. There are still places in the world where you can have access to free knowledge and courses. In my case it is because I work for a public institution. The course is in English, French and Spanish. This training helps you from constructing the problem statement up to the processing and interpretation of data, and including the design and conduct of semi-structured interviews.

It is a question of words and traditions. In quantitative classical approaches 'variables' are measured in numeric terms, and you use to build 'indicators' and 'indexes' representing theoretical 'dimensions' of a supposed more abstract 'concept' (I mean development, poverty, happiness, etc.). In classical qualitative investigation some of the most common terms were 'codes' and 'categories', but also representing 'dimensions' of abstract concepts or ideas to measure. Now all this terms stay, but there is some more flexible consideration of things when you are used of combining, mixing or triangulating. This is very clear with qualitative software when you can 'export' your codes or families of codes to Spss (for instance), and test relation of codes, etc.

I think Dean was right to draw the two terms 'abstract' and 'concrete' together to illustrate the differences between the two. In my view, it is all about 'boundary' or how we define things. In qualitative research, a researcher presents a 'full picture of a phenomenon' as a holistic or consolidated element. Different interrelated elements/information constitutes the broad 'phenomenon' of which boundaries are illusive. Whereas in quantitative research, 'variables' are distinct, measurable or countable. The broad phenomenon dealt as a single element in qualitative research is often broken down into several variables in quantitative research.

One purpose of qualitative research is to identify variables. Researchers begin with raw observations, and move from there to categories and preliminary concepts, and from there to potentially useful variables. Subsequent quantitative research establishes whether or not the variables are useful.

I agree with Sonya Jakubec and Hussin Hejase. Qualitative research has no variables. In qualitative research raw data are transformed into final description, or themes and categories.

Hi everybody, I see that there is no reason to say that qualitative studies cannot be based on identify and exam the impact of independent variables on the dependent variable chosen. A clear example in the qualitative methodology is QCA (Qualitative Comparative Analysis) which is based on the interaction among different independent variables in order to reveal an outcome in the dependent variable.

Ruben - not sure of your sources - but your interpretation is a very new one on me. Qualitative research does not work on the predictive relationship between dependent and independent variables

One of the main aims in qualitative research still is inference, and the logic to explain causal relationships (not the likelyhood for prediction of statistical studies) is based on the relationship between variables. QCA is a very clear example.

Hi Ruben

Maybe you need to clarify the way you address your point. I checked some recent papers and found an interesting one on causal research using qualitative approach.

http://dx.doi.org/10.2139/ssrn.1439868

Regards

Dear Hussin,

I will check this paper, thanks for the recommendation.

Best wishes,

Dear Amrit,

KIndly see an article on

"Quantitative Research Design" by Will G Hopkins

http://www.sportsci.org/jour/0001/wghdesign.html

Whether or not Qualitative Comparative Analysis (QCA) as advocated by Charles Ragin is or is not truly "qualitative" is a good question.

It certainly does turn qualitative data into "present-absent" (or 1-0) variables, so for anyone who objects to any kind of conversion of qualitative data into variables, it would not be qualitative.

But a different way to think about it is that QCA searches for patterns in the data, which then need to interpreted to be meaningful. So, if you are willing to dichotomize aspects of your data (i.e., turn them into variables), then can be a potentially useful too,.

Yes Hussin and David - looks like Ruben voted me down on my response. I admit, there can be some 'blurring' of qualitative versus quantitative variables - but it is very uncommon and often abused and, to me, relates more to mixed methods i.e Q- methodology approaches. Qualitatively - it all started with Strauss and Glaser's grounded theory - but their conventions are often mis-interpreted - as well as the fact they have continued to 'adapt' their methodology.

What is the relationship between qualitative research and variables? Are they similar to variables of qualitative researches?

In quantitative research hypotheses and variables are tested and measured numerically using statistical methods after the data were collected. Hypotheses and variables exist and stated prior to the beginning of investigation.

In qualitative research no hypotheses or relationships of variables are tested. Because variables must be defined numerically in hypothesis-testing research, they cannot reflect subjective experience. This leads to hypothesis-generating research using the grounded theory method to study subjective experience directly. The grounded theory method uses a data analysis procedure called theoretical coding to develop hypotheses based on what the research participants say. Because the method involves developing hypotheses after the data are collected, it is called hypothesis-generating research rather than hypothesis-testing research. Grounded theory method uses questioning rather than measuring and generating hypotheses using theoretical coding.

Are variables of quantitative qualitative researches similar?

The quantitative research leads to hypothesis-testing research (hypothesis are tested), whereas the qualitative approach leads to hypothesis-generating research (hypotheses are generated).

Hi Florencia your points make sense and worth investigating. After all, if we define very well our objectives and our expectations, mix methods will be helpful. Hypothesis generation is a constructive track to pursue and helpful to later on introduce quantitative analysis to measure. However, the researcher has to be careful to define the different steps and how mixing methodologies will proceed.

I think some of the debate here is between classic "inferential" statistics that are used to test hypotheses, versus a less well known set of quantitative tools that are used to search for patterns in the data. Since these techniques do not requires hypotheses (and do not provide statistics for hypothesis testing), they are usually labelled "exploratory."

Besides Qualitative Comparative Analysis (which is based on set theory), two other methods in this group are Cluster Analysis and Multi-Dimensional Scaling or MDS. Interestingly, some of the qualitative software programs (e.g., NVivo) now include Cluster Analysis as an option, based on overlapping codes in the "co-occurence matrix" (QDA MIner MDS as well).

Of course, doing this kind of analysis still presumes that you are willing to generate operational definitions for variables, even if you aren't doing traditional hypothesis testing. So, it is definitely quantitative in the sense that you have to believe there is something out there that can be measured and counted in some meaningful way.

Florencia writes ' The grounded theory method uses a data analysis procedure called theoretical coding to develop hypotheses based on what the research participants say'. Like so many GT people, she focuses on analysing what people SAY using interviews or focus groups. Yet why not work with naturalistic data rather than manufacture it through these methods? By studying behaviour [what people DO] , we can access phenomena unavailable to quant researchers. By contrast, being fascinated with what Florencia calls 'subjective experience' is precisely the territory of Oprah Winfrey and the psy professions. This point is developed in Jonathan Potter's demand that our data pass the 'Dead Social Scientist Test' and by Chapter 2 of my Very Short Book http://www.uk.sagepub.com/books/Book238751?siteId=sage-uk&prodTypes=any&q=david+silverman&fs=1#tabview=title

This is very interesting exchange of ideas. Colleagues are well appreciated for generosity in sharing their expertise.Personally, in simple way of appreciating variables and hypothesis in qualitative and quantitative research is that if variables and hypothesis are not quantifiable statistically because of subjectivity but investigation is imperative by exploring, describing, understanding subjective experience, event, culture will generate new knowledge and theory that will enhance how we view the world this will drive us to become qualitative researchers.

I second your appreciation Florencia. A continuous learning experience. By the way David, I am currently using Cluster Analysis in a paper in process.

Hi Colleagues, the discussion in this thread was quite vivid six years ago. Has any new dimension been added in the recent years to the field of knowledge related to the issue that I have raised in this question? If yes, please share.

A simple answer. Thinking about variables depends on a quantitative research design where we try to correlate operationally defined variables. By contrast, in Qualitative research, we investigate the very phenomena presumed in those definitions. Quant research cannot access the black box of how social phenomena are put together by participants in particular settings.

Note this has nothing to do with 'subjectivity' or studying 'experience'. Both of these are much more the concern of users of social media and the psy professions.

Just forget the dichotomy. Yes, the variables...same bits of concept. So is the space (the phenomenon of enquiry). This qualitative/quantitative dichotomy is a creation of egomaniacs of scholarship. Remove that dichotomy and we might see a better world. Look at the phenomenon. Tell us your lens and tools, and the vantage point. If the research is good, a few incisive questions will emerge. That is all that matters.