Quantitative Research Studies: An Introduction
Considering publishing a paper in the social sciences? If so, you need to be well versed in quantitative research, the type of research that for the last half-century has been preferred by most social science journals.
If you need a refresher or introduction, this primer offers a refresher on the purpose, philosophy, design, and methods behind qunatitative research studies, which represent the more general approach to scientific inquiry when compared to their counterpart, qualitative research.
Defining Quantitative Research
Educational psychologist John W. Creswell defines a quantitative study as “an inquiry into a social or human problem based on testing a theory composed of variables, measured with numbers, and analyzed with statistical procedures, in order to determine whether the predictive generalizations of the theory hold true.”[i] There’s a lot going on in that succinct definition, so let’s unpack it.
Quantitative studies apply the scientific method—historically associated with mathematical sciences like physics and chemistry—to fields where information isn’t always intuitively quantifiable. A researcher proposes a hypothesis at the beginning of their study and a quantitative study tests that hypothesis. A quantitative study assigns or measures numerical values for independent and dependent variables and looks for relationships between them. In many fields, like psychology, sociology, and biology, these variables are measured from a sample of subjects. Finally, the strength of the relationship between variables is determined with statistics.
A fundamental assumption of quantitative studies is that the researcher is an outside, impartial observer removed from the subjects and data of the study; he or she pursues an objective truth through a structured methodology that uses mathematical analysis. Some studies further guarantee against researcher bias by employing focus group transcription. Quantitative studies blur the boundaries between the so-called hard and soft sciences.They also set themselves apart from qualitative studies.
What’s the Difference Between Quantitative and Qualitative?
The qualitative mode of inquiry sees reality as socially constructed, with whatever meaning that might be teased out of a study to be found in the participants’ own words as recorded by a researcher, whose conclusions cannot by necessity advance beyond theory.
In qualitative research, you don’t test a theory; you aim to arrive at one. Witness the plethora of academic book titles that contain the phrase “towards a theory.” Within an academic field, the frequency of such books increases with the field’s reliance on qualitative research.
Historical Trends in Research
If the last half-century saw qualitative studies fall out of favor with social scientists, it also saw the humanities pick up some of that slack and embrace tools that once belonged mainly to social scientists. For example, look at the rise of cultural studies programs that are usually governed by English departments and whose chief concerns lie somewhere near the intersection of sociology, history, literature, linguistics, and philosophy.
A primary mode of investigation in cultural studies is ethnography, through which a researcher develops a theory of sociological meaning by observing and recording events and interviews within a defined population or space. Today, many ethnographies are carried out by academics in the humanities, like English professors and their students.
As defined here, contemporary qualitative research is informed by poststructuralist thought, and seeks a different type of control than the control groups of quantitative research provide. In qualitative studies, the researcher strives for what is perhaps the impossible goal of controlling for ideology; the qualitative perspective sees the primacy of numbers as a social construct, an ideology unto itself. Academic research transcription helps these intellectuals compile necessary information while preserving time for analysis.
Types of Quantitative Studies: Descriptive and Experimental
Whether or not objective truth exists, quantitative research gets us to a place that most can agree is at least pretty close to it through rigor in design and method. Two types of quantitative studies exist: descriptive (or observational or survey) studies and experimental (or longitudinal or repeated-measure) studies. In a descriptive study, the researcher does not intervene to change behavior or conditions. He or she simply measures things as they are.
Descriptive studies look for relationships, plain and simple. No causality is deduced, because these studies’ designs aren’t complex enough to know which variable caused the other one to change; the researcher just knows that they changed together.
An experimental study, on the other hand, has the researcher alter something about the subjects after an initial measurement. If something about the dependent variable changes after altering the independent variable, conclusions about causality can be drawn if all other variables are controlled for. Controlling for all relevant variables except the ones being measured is, of course, one of the difficulties of creating a good experiment.
A Comparison of Descriptive and Experimental Studies
Let’s consider a simple descriptive study as an example and then consider ways a similar study could be performed using anexperimental design. Take the age-old question of the relationship between classical music and cognitive performance.
A case-control descriptive study of the topic might look at two groups, listeners and non-listeners. What we would most likelyfind is that listeners perform better on cognitive tests than non-listeners; however, no causal evidence for that relationship exists on the basis of our simple measurement. Listeners of classical music constitute a select group that contains individuals likely to perform better on cognitive tests for reasons other than music preference. Classical music listeners are likely to be from wealthier families than non-listeners and possess more advanced degrees; to make a causal claim would be confounding the relationship of this third variable, wealth, with the music choice variable. Indeed, differences in cultural exposure make social class a common confounding variable in questions of test performance.
Now, suppose you take a random sample of the population or a stratified sample (a method of sampling to ensure that you have proportional representation of population subgroups), split the sample down the middle such that each half is effectively identical, and subject each group to both a treatment of classical music and no music (in the opposite order for each group and with a time gap sufficient to allow treatment effects wash out).
If you observe a relationship between music and cognitive ability, you’d have license to make a statement about causality, because you just used a crossover experimental design that controlled for time effects and selection bias.
How many measurements should a quantitative study make to draw generalizable conclusions? Usually, this question pertains to your sample size, which is usually determined via statistical significance. You want a sample size that’s large enough to detect a statistically significant effect between your variables 80 percent of the time.
Detecting a statistically significant effect means that 95 percentof the time (or 99 percent, depending on the significance level you’re looking for), you’d expect to see a smaller value than what you got for the effect you measured if the effect did not exist in the population. In the language of statistics, the p (for probability) value for the effect has to be less than .05 (or .01), the agreed-upon value that indicates that your observation was not due to chance.[ii]Naturally, the larger your sample size is, the more applicable your findings are for the general population.
Presenting Your Work
Because of their focus on conveying results that are replicable, the published forms of quantitative studies usually adhere to a predictable format. The body of the paper beginswith an introduction written in the third person—remember, these studies seek distance between observer and observed, in the name of objectivity.
As with any article, it’s the writer’s job in the intro to establish exigence – why we need their paper.The quantitative study does this byintroducing the problem it investigates, its variables and hypothesis (written as null and alternate hypotheses), the theories that inform the work, its key terms, and any limitations in its generalizability or replicability. From there, the researcher situates their study in the larger context of their field via a literature review. Here, a benchmark is established for comparing the results of the study with other findings. A methods section follows that contains descriptions of the research design, the sample or subjects used along with the researcher’s instruments and materials, and the data analysis of the findings. A concluding section is customary in which the author suggests ways to build on the work in question.[iii] APA style is usually preferred in journals that publish quantitative studies.
In conclusion, if you want an objective portrayal of a well-defined problem and you value replicability, control, generalizability, and explanations of causality, you should conduct a quantitative study to answer any and all of your burning questions. Consider using academic research transcription to ensure accuracy and objectivity.
Educational researcher L.R. Gay emphasizes what is perhaps the greatest advancement offered by quantitative studies: [E]xperimental research is the only type of research that can truly test hypotheses concerning cause-and-effect relationships.”[iv]
If you’re not sure that any amount of statistical analysis could get to the heart of your problems, you’d be better served with a qualitative approach. In truth, the best of both worlds can usually be had with some combination of the two techniques. These combination studies are known as mixed methods studies, and it’s towards them that the content of many humanities and social science journals is trending.