Collecting Meaningful Data

On the floor of just about any automotive manufacturing company, employees are collecting data and recording that data on a chart every hour as part of a program called Statistical Process Control (SPC) and once or twice a day, a floor supervisor checks those recordings for reliability and accuracy and once a day they are given to the Quality Department inspect for trends. 

 If any hourly measurement goes outside of what are called control limits, the process is immediately checked to determine what went wrong and whatever did happen, can be eliminated to the point that it never returns again.

Using acceptable standards for statistical interpretation, a supervisor can detect immediately if and when an employee is providing inaccurate data simply by counting the number of dots that have been recorded within a space and if it exceeds what statistically should be there, then the data has been recorded inaccurately.

And, while this process of checking quality is accurate for the manufacturing industries and some non manufacturing industries, what about the accuracy of verbal comments when it comes to measuring customer satisfaction?

For example, when questioned about their experience at a restaurant, customers reply that the experience was fine or acceptable when in fact it was not because they do not want to take the time to explain why nor do they want to get anyone fired... so, they provide the researcher with a misleading answer.

And, just because we have nothing better to do at the moment, let's suppose that these kinds of inaccurate statements are being collected by researchers all over America in a variety of questionnaires about a variety of situations... like a collective consciousness... although, how would you ever prove something like that?

Well...

How many Americans do you think fudge data on their income taxes each year?

How many Americans do you think exceed the speed limits when driving a vehicle?

How many Americans talk or text on their cell phones while driving a vehicle?

How many Americans steal pens and paper from work or us the copy machine for personal use?

There is a collective consciousness in American and all over the world, in fact, where people do similar things or say similar things without ever having discussed the issue and decided that they should act this or that way. Or, maybe it is just predictable human behavior given a variety of situations and stimuli. Or, maybe it is because of something else that we just cannot prove right now.

So, when marriages fail, can we really say for certain that the reason for that failure as articulated by both parties is accurate?

When a student drops out of college and is asked the reason for his/her leaving can we really say for certain that the reason given to us is really accurate?

These types of questions can be applied to literally thousands of research projects where data is being collected from which conclusions are being drawn as to why people do what they do, given a set of circumstances. And, these questions are asked in many different ways in the hopes of ruling out false data so that only accurate data is left and this is not doubt accomplished through various correlations of the data... and, providing those correlations are accurate, then one might be able to assume that the data collected is accurate and the assumptions extrapolated as well are accurate in a relative way.

In the Hawthorne Experiment, researchers were convinced that productivity was increasing due to changes in light levels because the workers had collectively concluded that management was paying attention to them in a positive way, hence changing the light... Therefore, in business schools, this scenario is shared with business students as proof that when management pays attention to the worker that output improves...

But, what if these results were misleading.... for instance, when anyone or everyone is driving on interstate highways, most everyone exceeds the speed limit and yet, then there is a highway patrolman present everyone seems to obey the speed limit... so, is this just another example of management being present? Or, is it an example of fear being induced into the situation with the outcome being a speeding ticket.

The Hawthorne Experiment could mean nothing more than the workers being induced with fear of losing their jobs since the lights had changed, they knew with reasonable certainty but not absolute that management was watching them.... and, when the light levels kept changing, then it was pretty much certain that management was watching them... and, they had no reason to believe otherwise or that an experiment was being conducted.

Therefore, the data collected is meaningful or meaningless?

Simply because PhDs collect data, perform research, analyze data, and draw conclusions so that they can write books or support dissertations, that we, the reader, should simply validate their conclusions once we consider the source... and, do all of our opinion have to be supported by what others think or by a preponderance of the data?

I teach a certain way in a classroom and get a fairly positive response from one or two out of 30 and so I must conclude that what I just did in class should never be done again because of the data that I just collected? Or, lets suppose I collected this same data in this same class every time it met for 15 weeks, do I now throw out that teaching technique because it did not work in this particular class? Or, do I need to keep collecting data for 10 years but on just this class or on all my classes... and, should I think about collecting data at other schools... and if so, from how many other schools? And, can other people collect data for me... or has that data collection got to be done by me to insure accuracy, reliability, and consistency?

When do I know that my data is meaningful or meaningless?

The United States currently has a divorce rate of 50% and some say our divorce rate is between 40-50% and some say is not even close to that number that it is more like 10%... and, if we look into the Baptist Bible Belt in the Southern US, we also find the same type of data: 50% or 40-50% or less than 10%... We have also (or at least some of us) read data that indicates that 4 out of 5 married males and 3 out of 5 married females have admitted to have extramarital affairs during their marriage. Of course, some of the males and/or females could be lying to boost their own egos.

How do we measure the happiness level that exists inside a marriage when there is no threat of divorce taking place? And, if by chance that happiness level is low... is that good or bad? If that happiness level is high... is that good or bad?

I cannot speak for all marriages but my parents were married for 50+ years before my father died and what I observed between those two partners on a regular basis was less than positive and this was also seen by my brother and sister as well. Our father drank too much alcohol and chased too many women but there is no direct evidence to support an affair... although, he and bridesmaid were discovered nude in a coat closet during a wedding reception once. Our father talked down to our mother and ridiculed her all the time while at the same time wanted us to show her respect. Our mother and father took separate vacations after the 3 of us graduated from high school. Late night conversations were heard between our father and mother regarding his adultery and several times our mother told us that the two of them were separating but that never happened.

Our dad had a favorite saying: “If you act like you are doing nothing wrong, most people will think you are doing nothing wrong.”

I realize that this is limited evidence and that the two primary parties in question were never interviewed but is it possible for one to conclude that this was not a happy marriage and yet the marriage remain intact for whatever reason(s).

I would have to say that 80-90% of the people with whom I attended high school have shared similar stories with me about their parents. But again, is this sufficient data on which to make the statement that we now have meaningful data?

Can we conclude that meaningful data can only be based upon the statistical random but systematic collection of raw data? If so, then what does our divorce rate conclude about those couple that have decided to remain together?