# Lying With Statistics

Posted on

Data can easily be skewed in order to prove or disprove a point or area of research. When looking at progress towards the Sheff I goal, information portrayed in a graph can be situated to look like schools are making either grand leaps in progress, or small steps. After taking data from Figure 5.1, titled “Actual and Legal Progress toward Sheff I Goal, 2003-2007″, and making it into a line graph, it is easy to see that data can be skewed in order to prove a point.

Slow Progress Toward Sheff I Goals

Looking at the graph to the right, it looks like schools are not making a lot of progress in regards to meeting the 2006-07 goal. The slope of the line is not very steep because the y-values are separated by increments of 20%. This makes the progress line seem horizontal and slow-moving. It looks like it will take some time before schools meet their goal of 30%. I made this graph by utilizing the “Insert Chart” option on Google Spreadsheet. By fiddling with the minimum and maximum values, along with the scale of the y-axis, I made the graph look like schools are not making progress towards meeting the 2006-07 Sheff I goals.

Quick Progress Toward Sheff I Goals

Looking at the graph to the left, one would assume that the schools are making steady progress towards meeting the 2006-07 goals. I purposely did not include the “Goal” line of 30% integration because I wanted to make the line as steep as possible (within reason). In order to do this, I made the y-interval scale .1 and only showed five percentage points. Since the y-interval maximum was 14%, the line goes beyond the graph, which makes it seem like schools are making great progress. If one were to glance at this map without looking closely at the y-intervals, one would assume that schools are making progress.

I found this exercise very helpful in terms of understanding how people can play with graphs in order to prove their point.