Data can enlighten, or deceive


As school officials continue to monitor the COVID-19 situation with plans to open schools Aug. 21, a number of statistical factors will no doubt be used as part of the process.

But to paraphrase Mark Twain’s famous quote, statistics cannot always be believed, and they can be manipulated to argue one’s case in either direction.

The numbers coming from the Department of Public Health (DPH) and other healthcare professionals have been used in articles all over the state to justify plans of action ranging across the entire spectrum about whether schools should return to in-person learning.

At the July 20 school board meeting, Superintendent Matthew Cooper cited the DPH’s data and quoted a 7-day moving average of COVID-19 cases in Habersham County at 1.9, saying it was “the lowest it has been since the virus came (here).”

Let’s remember, there is a lag in the calculation of the data, as cases are processed in different areas depending on where someone went for treatment of the disease.

A reader pointed out to both The Northeast Georgian and Cooper that the stats must be taken with a grain of salt in light of these factors.

If you look at DPH data for the last seven days (as of Tuesday’s report), the total number of cases since February has increased to 961, with 85 cases added to that total over the last week. That comes out to a 7-day moving average around 12.3 cases per day.

If you look at the last solid data point from July 13 – DPH recommends reading it two weeks out – the number was 9.3, still quite a bit higher than 1.9.

“This current set of data he has shared is not relevant to actually going back to school on Aug. 21. We will be evaluating all available data when we get a couple weeks out,” Cooper said Wednesday. “It is way too early to think that the data we have now can inform our decisions in late August.”

Cooper has stressed that his intent in the last board meeting was “to get people to start looking at other data and not just the total number of positive cases since February.”

Indeed, there are 961 cases reported in Habersham County in that time, but there are far fewer active cases now.

Even the total deaths from COVID-19 in Habersham County – 48 as of Tuesday – need to be interpreted with some deference to factors such as the two nursing homes in the county seeing 37 deaths alone. Again, not all the residents of those homes are legal Habersham residents, so even that number is not set in concrete for how it should be viewed.

So knowledge can be gained by viewing the statistics, but they cannot provide total certainty about a virus this unpredictable.

Cooper added that he plans to continue focusing on figures from official guidance and interpretation from state health officials, not political motivations, when making decisions about the safety of students.