Once you get started with human behavior research you soon find yourself running into the question of whether your research project is qualitative or quantitative in nature. There are inherent differences between qualitative and quantitative research methods, although their objectives and applications overlap in many ways.
The core difference
In a nutshell, qualitative research generates “textual data” (non-numerical). Quantitative research, on the contrary, produces “numerical data” or information that can be converted into numbers.
Qualitative research is considered to be particularly suitable for exploratory research (e.g. during the pilot stage of a research project, for example). It is primarily used to discover and gain an in-depth understanding of individual experiences, thoughts, opinions, and trends, and to dig deeper into the problem at hand.
The data collection toolkit of a qualitative researcher is quite versatile, ranging from completely unstructured to semi-structured techniques.
Most common applied Qualitative Methods:
In addition, eye tracking or automatic facial expressions can be collected and analyzed qualitatively, for example in usability research, where gaze patterns (such as with heatmaps) or moments of expressions of frustration / confusion can be used to track the journey of an individual respondent within a software interface.
Typically, qualitative research focuses on individual cases and their subjective impressions. This requires an iterative study design – data collection and research questions are adjusted according to what is learned.
Often, qualitative projects are done with few respondents and are supposed to provide insights into the setting of a problem, serving as a source of inspiration to generate hypotheses for subsequent quantitative projects.
Simply put, quantitative research is all about numbers and figures. It is used to quantify opinions, attitudes, behaviors, and other defined variables with the goal to support or refute hypotheses about a specific phenomenon, and potentially contextualize the results from the study sample in a wider population (or specific groups).
As quantitative research explicitly specifies what is measured and how it is measured in order to uncover patterns in – for example – behavior, motivation, emotion, and cognition, quantitative data collection is considered to be much more structured than qualitative methods.
Quantitative research techniques
Quantitative techniques typically comprise various forms of questionnaires and surveys, structured interviews as well as a behavioral observation based on explicit coding and categorization schemes.
In addition to these traditional techniques, biosensor recordings such as eye tracking, EEG, EDA / GSR, EMG, and ECG as well as computer-guided automatic facial expression analysis procedures are used.
All of these quantify the behavioral processes in such a way that numerical results can be obtained – for example, fixation duration from eye tracking (representing the amount of visual attention), the number of GSR peaks (indicating the amount of physiological arousal) or the power of a specific EEG band.
After data collection, quantitative analysis techniques and statistics can be applied, such as t-tests and ANOVAs, to non-parametric methods. This often necessitates much bigger sample sizes compared to qualitative research but allows you to make more solid conclusions, that are backed up with data.
Ultimately, whether to pursue a qualitative or a quantitative study approach is up to you – however, be sure to base your decision on the nature of your project and the kind of information you seek in the context of your study and the resources available to you. Qualitative will offer you an in-depth understanding of your research problem and hopefully help answer your hypothesis. Quantitative will allow you to scale your research to provide larger sets of data for reliability and validity. A combination of the two provides you with objectivity.
This is generally described with respect to the following criteria:
Objectivity is the most general requirement and reflects the fact that measures should come to the same result no matter who is using them. Also, they should generate the same outcomes independent of the outside influences. For example, a multiple-choice personality questionnaire or survey is objective if it returns the same score irrelevant of whether the participant responds verbally or in written form. Further, the result should be independent of the knowledge or attitude of the experimenter, so that the results are purely driven by the performance of the respondent.
A measure is said to have a high reliability if it returns the same value under consistent conditions. There are several sub-categories of reliability. For example, “retest reliability” describes the stability of a measure over time, “inter-rater reliability” reflects the amount to which different experimenters give consistent estimates of the same behavior, while “split-half reliability” breaks a test into two and examines to what extent the two halves generate identical results.
This is the final and most crucial criterion. It reflects the extent to which a measure collects what it is supposed to collect. Imagine an experiment where body size is collected to measure its relationship with happiness. Obviously, the measure is both objective and reliable (body size measures are quite consistent irrespective of the person taking the measurement) but it is truly a poor measure with respect to its construct validity (i.e., its capability to truly capture the underlying variable) for happiness.
If you would like to learn more about qualitative and quantitative research designs, contact our experts at iMotions. We’re happy to help!