Sunday, June 14, 2009

In Which I discuss relevant data

During my tenure in the QA department, I worked with a lot of data.

Excel and I, we were tight. Charts and graphs, trend lines and correlations... man, I was in my prime.

All of this, taught me a thing or two about data, and what a trend actually is. A trend, and the period of time that constitutes a trend to ensure you have enough data to look at. Also, I learned what an anomaly is, and what needs to be looked at a little closer.

I got an email from my boss telling me that I'm preforming poorly simply because over the last 6 shifts 101 of my records took longer than the target amount of time to complete. In these instances, I went over twice the amount allotted.

Now, this sounds bad... at first. When you consider the fact that in those 6 shifts, I handled well over 2500 records. This would mean that at least 2399 of my records were handled well within parameters.

There are two things that I have to say about my boss saying that based on this data, I am trending quite poorly.

The first is the sample isn't large enough to really demonstrate that this is a trend. 6 days, just isn't enough raw data to work with. Add in another week, and maybe then we're talking.

The second, based on the above information, this is only 4% of the time that I am out of parameters. This, in my opinion, having handled a department where efficiency was key, is cause for a warning... not a veiled threat to my employment such as I received from my employer.

Needless to say I am very frustrated at present, but I'm doing what I can to take this in stride. Sadly though, I'm not terribly thick skinned. This coupled with the fact that the only feedback I get is when I've done something wrong... is really wearing on me.

No comments: