Don't get me wrong! I know the differences between qualitative and quantitative research.

I know we need sometimes to measure and other times to understand, I know sometimes we test hypotheses using statistical tools and other times we don't. I do not want to eliminate any possibilities and approaches we have in social research. I don't want to impose objectivity when it is not possible, I don't want to forget about context when we need to consider it and so on.

I'm just saying that we imposed to ourselves some rigid boundaries like this is a qualitative/quantitative research. And after we accepted that we can and sometimes even must combine them, we started imposing other boundaries like this is sequential, concurrent and other qual-quant or quant-qual combinations.

I find it useless and dogmatic. Let us combine anything we want the way we want when is possible! Perform qualitative analysis on open ended questions, count word frequency in interviews, test on larger populations hypotheses resulted from a small sample size and others like this without worrying about what kind of methodological approach it will be! It is social research!

It is about methodological liberty and creativity and about making hard things easier. But it is also about accuracy. Most of the social data are qualitative (even in "quantitative"studies); most of our studies require both "qualitative" (even it is just thematic analysis) and "quantitative" analysis (even it is just counting); we use both induction (even in "quantitative" research) and deduction (even in "qualitative"research) in every social research. I think we employ improperly qualitative and quantitative terms. For qualitative and quantitative analysis, for example, we have other terms matching perfectly like coding, content analysis etc.; for qualitative interview we have accurate terms like unstructured or semi-structured interview and so on.

Anyway thank you all very much for your interest in my question! It is a wonderful community feeling that your answers bring to me even it looks like I'm a lone rider.

Hi Florentina,

Your evaluation is partly correct. The difference between quantitative and qualitative data is essentially that between numerical and narrative data sets. How the data is collected in survey's depends to what extent the questions are closed, open and/or semi-structured. Since the advent of mixed methods approaches we can. thankfully, mix them all and view the world in the 'bigger picture' context. 

The attached chapter may assist. 

Dear professor Whitehead,

Thank you very much for your reply and for your paper!

I think being structured does not make data quantitative. I think a closed question about an opinion, for example, still brings qualitative data. Of course structured data are easier to be quantified and this is what we do with them in quantitative researches. But being structured or not, they are in their primary form qualitative data. I think this is why we also call these variables - categorical.

I would agree with you. Quantitative Date is anything expressed with numbers. Qualitative Data anything that is not expressed through numbers.

However, there are different understandings among disciplines. I.e. a friend in Biotech said, that they consider everything qualitative, they do not measure with appropriate tools. My professors in engineering considered everything qualitative that originates from questionnaires. So hence, I suggest to consider the audience whenever using the terms and possibly defining them.

Some further thoughts on your examples: In natural sciences (and some social sciences too), quantitative data has a higher "value" than qualitative data. Therefore there are some ways to transfer the qualitative into quantitative data, i.e. Dichotomization, Word counts, Image Histograms, Counting. So your question might finally yield numerical data, notwithstanding what you asked in the questionnaire.

The other way around is possible too, from quantitative to qualitative data (i.e. by writing out the number: 1000 becomes "one thousand" and eventually is even ornamentally drawn. In such cases, the qualitative data contains more redundancy...

Thank you very much!

Very interesting your observation about different understandings among disciplines. Isn't it a strange situation? How can we talk about transdisciplinarity, multidisciplinary, interdisciplinarity if we have different understandings about the data we work with?

I think we have quantitative and qualitative data in every discipline and that they have the same definition.

I think it is hopeless to try to create an airtight categorization of data as either qualitative or quantitative, but we certainly know the difference between the typical form of quantitative research (statistical tests with numerical data) and the typical form of qualitative research (interpretive analysis of textual data).

Trying to dismiss the distinction between qualitative and quantitative research by picking on one disctinction after another (e.g., numerical versus textual data) just doesn't add up. I have attached a copy of my forthcoming article in the Journal of Mixed Methods Research, which discusses these issues.

Article Living Within Blurry Boundaries: The Value of Distinguishing...

I certainly agree with David L. Morgan in his argument that it makes sense to distinguish between quantitative and qualitative methods (and related techniques) and there is no sense in distinguishing between qualitative and quantitative sets of data.

However, I would add that the real argument for the "hopelessness" here is that you can certainly apply both types of methods on any data (depending on your research question) making the data effectively qualitative or quantitative. With some, both might make sense, while with others, all conceivable research questions lead to a sensible use of only one type. 

The key cerainly thus is not in text/number distinction, but more emphasis on the way data are dealt with. I think this could preclude inter/multi/trans/disciplinary in any way...

Data is Quantitative or Qualitative depending on its application by a research in relation to the discipline under focus. In actual sense, i think that it is the manner by which a researcher handles the data in his qualitative research that makes the data qualitative data or in his quantitative research, that makes the data quantitative data. Granted that  qualitative data can be converted to quantitative data and empirical data can be used to conduct qualitative research. The approach therefore seems to be the deciding factors.

I have two thoughts about using "only one type" of method per research question. On the one hand, I am an advocate of using mixed methods. On the other hand, I prefer to have the two methods serve separate goals in what are known as "sequential" designs within the field of mixed methods. That way, each method serves a specific purpose with regard to the other.

Within the field of mixed methods, one of the biggest issues is the "integration" of  the results from qualitative and quantitative studies. Achieving integration is straightforward in sequential designs, where the results from one method meet the needs of the other method. Where integration gets especially messy is if one fails to distinguish how the goal for the two studies will relate to each other, and then ends up with potentially contradictory results.

So, I think that successful mixed methods requires paying very careful attention to both the kinds of data you will collect and how you will use those different kinds of data.

Thank you all very much!

I know my place but I think I'm going to fight some more for my convictions!

I see things the other way around. We only need the distinction between the data type in order to find appropriate ways to collect, analyze and interpret them. For example for not measuring what could not be measured.

I do not think we really need the distinction between qualitative and quantitative research and I prefer the term triangulation and not mixed-method (I think the last idea is Uwe Flick's idea). You could do and in some cases you need to do data, method, investigator, theoretical triangulation in a research. You need some skills and you need to follow some rules in order to collect, analyze and interpret different kind of data and this is it. The research is one and you don't have the integration problem.

Qualitative approach dwells much on acquiring rich, detailed, 'thick' and 'quality' data on a phenomenon under study which requires extensive collection of particularly verbal and textual data from researchers who have had great engagement (time) with informants. Thus, it takes a great deal of time, patience and other resources in studying in detail the PERCEPTIONS and ATTITUDES of people regarding issues under investigation. Thus, these cannot ideally be undertaken using statistical paradigms. Thus, its data is efficiently gleaned via intrrviews and observations.

Quantitative approach dwells largely on statistical data and analytical tools to draw conclusions about the state of a phenomena under investigation. It is very appropriate when the researcher is engaged in experimental research that employs laboratory inquires rather than perceptions of persons to answer research questions.

Qualitative is said to be subjective while quantitative is objective though there are various ways of reducing the subjectivity of qualitative researches. 

Interestingly, using both means for a particular research yields the best results. The mixed method approach with its varying types must be studied by researchers so that they will know which of them best suites their research objectives. Of course, pure anthropological studies in many cases requires strict adherence to qualitative approaches though responses can be coded and translated into numerical values for quantification.

Personally, I think both approaches are not 100% in opposing sides of purely textual data (qualitative approach) and numerical data (quantitative data). Yet, the greater quantities of each of the streams of data as well as their analytical procedures used for analysing the accrued data gives the distinctions.

Thanks

Dickson Adom

Simply identified as Quantitative techniques (or positivist research) involves approaches in this sequence- theory-hypotheses-data gathering-analysis and back to theory. This technique allows for concepts to be duly ope-rationalized. It is a research technique that is grounded by focusing on numerical data based on the observation of behaviors of the selected sample size.

While Qualitative technique (or constructive research) lacks the precision and definition of the quantitative technique but has a specific data gathering method which involves direct observation, in depth interview and use of varied documents. The dominant methodology is to discover the meanings and interpretations of the collected information obtained from the subjects in their natural settings,

 However, these two varied techniques makes good a research topic when they are combined to create a Mixed method approach- a type of research in which quantitative and qualitative techniques are combined and used in a single investigation.

There is no ambiguity in the differences obtained in the aforementioned research techniques.

I'm kind of young (or let's say unexperienced) and I probably should not answer this but I hope you will correct me if I am wrong :)

I think It's been implicitly pointed out that in the end the answer for this question lies in the mixture of how you frame the phenomenon that interest you, what level of analysis you decide to apply and eventually what ontology and epistemology you choose to believe to be true (and therefore take it for granted). Of course if you're a pragmatist, you are also much more concerned with the praxis of explanations presumably resulting from the study or sequential design of studies - as pointed out by profesor Morgan.

I would also keep the distinction between quantitative and qualitative research rather than the data itself, because the difference is much more obvious when you look at the whole process. To name one example I believe both kinds of research usually take different spin on the implicit attitude towards reductionism. In quantitative research we use numbers to reduce the complexity of phenomena willing to offer a clear, sound and generalizable explanations. Meanwhile in qualitative research it is not the aim at all. We spend weeks or months  digging into the data, trying to understand the complexity of phenomena as it is, without reducing it particualrity. You can see it when you compare how much time quantitative and qualitative analysis takes or how long the 'findings' sections in the papers are in both cases.

I guess what I am trying to say here is that in times of paradigmatical plurality those lines got to get blurry. But it's good because it make us ask questions like yours and wonder how to link things we believe to be true and (most importantly) care about others to convince them what we do makes sense, without assuming everbody ought to think in the same way as we do. As pointed out by Beni Rohrbach you can change qualitative into quantitative data and vice versa all you want, however as suggested by professor Morgan, the real question is what reason you have to do so in light of the method you apply and the research question you ask.

I share the article you might like. It is kind of liberating but in a way that gives you a lot to think about :)

To distinguish quantitative and qualitative research is a well-known tradition in empirical sciences. There exists the classic field of quantitative research like hypothesis testing or to describe areas or events with relevant percentages. On the other hand there is the field of qualitative research with partly high sophicated methods, for instance in (empirical) biographical research, or literary work (for example: to find out the trends of interpretation of a special problem within a certain time space and to collect various written articles in journals to determine a structure or a trend of interpretation), apart from historical subject-matters. It is right that scale theory has both qualitative aspects (so yes-/no-answers) and quantitve aspects (so gaining a normal distribution of data), but scale-theory for itself belongs to quantttatve methods, of course. The ideal of science is to rule nature more and more by predicting effects of variables and events on the base of substantive theories. The last aim is to develop techniques and technolgy in manipulating sometihng for a useful reason. Qalitative research often follow another end: not only to rule the world - but to understand  men and the university! Tthat's another aspect of  human endeavour. The simple answer of your question is: If you have to do with numbers in a quantifying manner than you have to do with quantitative research; quantifying means to allocate numbers to events. But not all kinds of research make sense to do this. Perhaps you know  the well-known statement: Quantitative working researchers get there results very preciserly, but what they quantify rearly they don't know. Vice versa: Qualitative researchers may understand the world but they can't improve the scientific-technical progress to better the situation of society. At last both views are reasonable and need ech other. The quantitatve researcher uses phrases, he needs the language, qualitive variables, in order to understand and to discuss the situation which motivates him to construe a substantial hypothesis. And he has to understand and to interprete the result of his testing. So both methods  work together. In my personal view it is Important is that students of social sciences learn to apply a broad field of quantative and qualitative methods.

With regard to triangulation, the earliest approach was to compare the results to see if they were equivalent (QUAL = QUANT). Flick's work shifts this orientation to something like QUAL + QUANT, to see what can be gained through the combination of the two.

So, I would certainly see that as maintaining the general division between qualitative and quantitative data. More specifically, his empirical work has a strong tendency to identify processes as something that are best studied qualitatively and structures as something that are best studied quantitatively. His version of triangulation then works back and forth between those two kinds of data to find the best overall set of conclusions that combine the different perspectives.

I find that social data is almost always non-parametric/qualitative, or presents mixed methods requirements such as quantitative qualitative analyses of interviews, questionnaires, surveys, polls etc. Physical substrates really do qualify for quantitative measurement - e.g., blood cell counts etc. That said for efficiency purposes even laboratories round up by providing estimates based on subsets of substrate data for everyday purposes, e.g., DNA samples are presented as random-match-probabilities, so... I guess most data is statistical whether quantitative or qualitative.

Don't get me wrong! I know the differences between qualitative and quantitative research.

I know we need sometimes to measure and other times to understand, I know sometimes we test hypotheses using statistical tools and other times we don't. I do not want to eliminate any possibilities and approaches we have in social research. I don't want to impose objectivity when it is not possible, I don't want to forget about context when we need to consider it and so on.

I'm just saying that we imposed to ourselves some rigid boundaries like this is a qualitative/quantitative research. And after we accepted that we can and sometimes even must combine them, we started imposing other boundaries like this is sequential, concurrent and other qual-quant or quant-qual combinations.

I find it useless and dogmatic. Let us combine anything we want the way we want when is possible! Perform qualitative analysis on open ended questions, count word frequency in interviews, test on larger populations hypotheses resulted from a small sample size and others like this without worrying about what kind of methodological approach it will be! It is social research!

It is about methodological liberty and creativity and about making hard things easier. But it is also about accuracy. Most of the social data are qualitative (even in "quantitative"studies); most of our studies require both "qualitative" (even it is just thematic analysis) and "quantitative" analysis (even it is just counting); we use both induction (even in "quantitative" research) and deduction (even in "qualitative"research) in every social research. I think we employ improperly qualitative and quantitative terms. For qualitative and quantitative analysis, for example, we have other terms matching perfectly like coding, content analysis etc.; for qualitative interview we have accurate terms like unstructured or semi-structured interview and so on.

Anyway thank you all very much for your interest in my question! It is a wonderful community feeling that your answers bring to me even it looks like I'm a lone rider.

Good explanation, thanks for your contribution also and we appreciate you asking the question. 

I agree with your thoughts. This type of data is often confused because it is associated with the method used. Though the data type shows the type of data we need for our research. While the method in it contains among other ways to get various types of data. Both can exchange, that quantitative methods can also be used to obtain the type of qualitative data. Vice versa, qualitative methods can also be used to obtain the type of quantitative data. For example, I ask in depth about the income of farmers in rural areas. I use a qualitative approach to get a quantitative income. Instead, I used an open questionnaire to get a farmer's explanation of the farm they developed. Means I use quantitative methods to get qualitative data types. Such is my simple thought, thankyou.

Essentially, qualitative data is that which remains unquantified.

For instance, the response "No" to the question "Did you inject today?" can easily be quantified.

The response "Yes, and it was marvellous; all my troubles just vanished; I was also aware that I had wet myself, but I didn't care." to the same question can also be operationally quantified, but with loss of data.

Any response can be quantified, but possibly with data loss regarding the research context.

Thus, qualititative data is that which remains unquantified.

Qualitative data collection can be used to gather quantitative data however it might be in smaller scale in some research areas. But quantitative approaches such as survey will not allow the researcher to gather enough data. However it should be noted that the quantitative data can also have qualitative and descriptive interpretations. 

Qualitative data is a data concerned with descriptions, which can be observed but cannot be computed, quantitative data is the one that focuses on numbers and mathematical calculations, and can be calculated and computed. These data types are used in a number of fields like marketing, sociology, business, public health and so on.

I agree, quantitative and qualitative not just data types, but most importantly what is the basis of choice. I think the purpose of research becomes the first important basis, then proceed with how to answer the purpose of research, so that can be designed what tools will be used to obtain data, finally what the appropriate analysis tool used. Problem data types can vary, quantitative data can be used in qualitative research. Similarly, the qualitative data can also be used in quantitative research. But certainly speaking the obvious methodology are different. So my tool, hopefully useful. Thankyou

simple as that . Data generated on measurable characteristics are designated as quantitative data and data generated on qualitative characteristics(Non measurable . categorical) designated as qualitative data.