Question:

In spite of the fact that we normally associate quantitative approaches with numbers and “detached” analysis, all quantitive methods at some stage involve sociologically-informed decisions and subjective interpratation. In what ways is this so?

(make a reference to DAMNED LIES AND STATISTICS, AMONGS OTHERS). It is also important to give true examples to substantiate your arguements

Answer:

Title: Quantitative research methods in sociology

Introduction

We normally associate quantitative approaches with numbers and ‘detached’ analysis. However, at some point, all quantitative methods involve sociologically-informed decisions and subjective interpretation. There are numerous examples of how some aspects of quantitative analyses are full of subjective interpretation informed by the prevailing sociological contexts. This paper explores, with appropriate statistical examples, the different ways in which this happens.

How quantitative approaches are associated with numbers and “detached” analysis

For most people, statistics can be presented in ways that support a certain view or contention. To this extent, they are said to be presented in a subjective manner. A common manifestation of this subjective approach is using the median income and average income as if these two statistical terms mean the same thing mathematically. The median is different from the average in that it represents the ‘middle’ value in a certain set of data, which has been arranged from the smallest value to the highest value.

It is common for statistical information to be presented in a manner that ‘proves’ a point (Best, 2001). Ideally, statisticians ought to always think about every data set from which the statistical information has been derived. Unfortunately, laypeople lack access to this data set, meaning they tend to be easily swayed into accepting whatever opinions that the proven by the analysts Best, 2001). This explains why quantitative researchers, unlike the public, tend to disagree with each other on the outcomes of their surveys. The main difference is that the public tend to think in terms of counting while quantitative surveyors have to measure and to deal with analysis that is detached from sociological contexts.

Quantitative researchers understand that their measurements may not easily agree exactly with all the values reported by all other surveyors. The differences may occur as a result of many factors, including different equipment and procedures. For instance, whereas one researcher may be able to measure a given distance directly, an earlier survey might have necessitated the use of a multi-station traverse, also commonly known as indirect measurement, for measuring the same line.

The best way for researchers to differentiate between detached analysis and subjective sociological decisions taken during quantitative research is by comparing counting and measuring. In counting, the prime goal is to determine discrete quantities that are indivisible. For example, the number of people in a lecture room constitutes a whole number, which can be determined by simply counting heads (Fielding, 2001). If one can count in the proper manner, the counting process will be without any error, such that another person can repeat the count and end up with the same number. However, if scales were brought into the room for measuring the total weight of the people present, each scale would deliver a different answer (Fielding, 2001). This simple example demonstrates the difference between counting and measuring. Whereas counting is based on a sociological context, measuring involves only detached analysis.

Unlike counting, the measuring process involves estimation. Regardless of the precision of the instruments used for measuring, an estimate has to be made sooner or later, which definitely result in errors (Balnaves, 2001). Whenever estimation is done, subjective decisions are made, leading to errors. These subjective decisions are a result of sociologically formed decisions. In other words, our measurements are always estimates of the truth, and this makes the truth very hard to come by in this whole process.

Whether the errors that arise in the measuring process are systematic, random, or merely a blunder, sociological considerations always come into play in making decisions (Bulmer, 1999). Systematic errors can be computed, and subsequently, can be eliminated from the measuring process. However, random errors, which occur in the measuring system, always remain in place even when systematic errors and blunders have been eliminated. This is because they cannot be predicted with certainty, meaning that they are highly prone to our subjective sociological interpretations. On the other hand, blunders are simply mistakes, arising out of the imperfect nature of human actions, leading to erroneous values. They represent the greatest influence of sociological practices on the process of carrying out quantitative research.

How at some stage quantitative approaches involve sociologically-informed decisions and subjective interpretation

There is no single tool used for the survey of measurements that is perfect. This means that every quantitative approach contains some of error or uncertainty, no matter how small it may be. This room for uncertainty paves way for subjective interpretation. For example, steel tapes used for measuring length may expand or sag with changes in temperature (Best, 2001). Similarly, rag tapes may naturally stretch because of exposure to different environmental conditions. Moreover, operators may make imprecise readings or fail to get set-up perfectly in every situation, often because of sociological biases (Wolf, 1986).

Although errors are unavoidable in every measurement, blunders are foul-ups that must always be eliminated from all measurements (Wolf, 1986). Researchers have to avoid setting up on wrong points, sighting the wrong back-sights, or transposing numbers in field notes, all of which are examples of blunders. On the other hand, errors need to be controlled, mainly by keeping the equipment in good shape and adjustment all the time and using observing procedures that lead to the cancelling out of errors. Moreover, many instrument errors cancel out as a result of observing forward and reverse positions whenever angles are being measured or lines are being extended.

To some extent, systematic errors arise as a result of the observer’s physical limitations or biases. Specifically, three sources of systematic errors that are known to exist in any sociological quantitative research method include natural errors, instrumental errors, and personal errors. Natural errors are caused by environmental sources, of which no man can exert any control. Therefore, any corrective measures made arise largely from subjective interpretations of the degree of the prevailing environmental changes, for example, an increase in temperature. Similarly, instrumental errors lead to situations where the researcher has to rely on the established norms of dealing with the shortcomings caused by the imperfections of the instruments being used for obtaining observations. In personal errors, the observer may be biased either randomly or as a tendency to make observations in a certain way. Random bias is caused by the very physical limitations of all human beings, while the tendency to influence readings in a certain way may lead the observer to always ‘read high’, thus causing a skew in the data set.

Conclusion

In summary, in spite of the tendency to associate all quantitative approaches with numbers and the corresponding ‘detached’ analysis, it is impossible to avoid some sociologically-informed decisions and subjective interpretations at some stage in every quantitative method. Unlike the counting process, which laypeople easily understand and rarely dispute, the measuring process that quantitative researchers use encompass some aspect of estimation, since neither the instruments used in the estimation process nor the people using them are perfect. This explains why there are systematic errors, random errors, and blunders, and by extension, different interpretations of the same sociological phenomenon by different sociologists.

References

Balnaves, M. (2001) Introduction to quantitative research methods: an investigative approach, London: Sage Publications.

Best, J. (2001) Damned Lies and Statistics, Los Angeles: University of California Press.

Bulmer, M. (1999) Sociological Research Methods, London: Routledge.

Fielding, N. (2001) Introduction: On the Compatibility between Qualitative and Quantitative Research Methods, Quantitative Social Research, 2(1), 12-39.

Wolf, F. (1986) Meta-analysis: quantitative methods for research synthesis, London: Sage Publications.