Making Numbers Make Sense

My research beliefs completely changed when I took up qualitative research under a professor who is quite passionate and well-versed with the subject. I was fascinated with how qualitative researchers collect and interpret data. The “in-depth” analysis of data somewhat made quantitative research less credible. Example, a quantitative researcher might say the result is not significant if there is only one rape case per 10,000; however, the qualitative researcher will consider that lone case and declare the results to be “significant”. This example cited by my teacher keeps reverberating in my mind. After the end of the semester, I developed my utmost respect to qualitative researchers and I stopped cultivating my passion for quantitative research. I even started to think that qualitative research is better.

How about marrying the two types of research? I bow my head to the one who first thought of this idea. This would definitely settle the issue regarding the differences of quantitative research and qualitative research. With the “marriage” of the two researches becoming a success, researchers should be keen enough to polish their skills in analysis of narrative and numerical data.

For those who are still groping in the dark on how to interpret numerical data the PowerPoint presentation Making Sense of Numbers by Dr. Jim Adams-Berger would definitely help. Those who have learned the ropes in data analysis could use the presentation to recall the basic considerations on interpreting “numbers”.
The presentation encompasses using data to identify community problems, to explore problem relationships and to choose responsive programming. As a whole, Making Sense of Numbers is anchored on the framework of research (perhaps a research program particularly designed for OMNI Research and Training, Inc.). Berger points out that values serve as a filter of the other major analytic elements, which are beliefs, data and research. There is also a detailed explanation on modeling problem relationships. The use of graphic models makes the presentation more substantial and interesting especially for those who have just started doing research.

A wide range of examples could have been presented in surfacing and identifying problems; however, the presentation focused on the differences of numbers and rates in a particular case. In a certain example, the numbers of aggravated assaults for 4 consecutive years are not enough for us to conclude what year has the worst record. Calculating the rate is more meaningful. The recommended procedure in calculating the rate is number of cases divided by the population and multiplied by a large number (“per” number). That would be acceptable if the number to be multiplied is smaller than the population. What if it is larger than the population? One murder case in a population of 500 may be interpreted as 20 murder cases per 10,000. Is it not misleading? I think it would be better to stick to the usual per 100 in solving rates.

It is also best to consider different viewpoints before making conclusions from a set of data. There are instances when rates are favored than numbers, but there are cases when rates give misleading information.
In the example in calculating rates, the data in year 1997 shows population of 10,000 and number of aggravated assaults of 70. However, the rate per 10,000 is 7. Let us take a moment to ponder if it is correct. Perhaps, a calculator could help.
Aside from faulty calculations, another pitfall of “numbers” is the method of gathering data. There are a lot to consider – from the structure of questions to the respondents’ willingness to answer. That is, we have to ensure validity and reliability.

I consider myself a neophyte in the research arena. I realize there are still a lot to learn. Starting with Making Numbers Make Sense I hope to fill my brain with the skills in numerical data analysis that are useful in research.