Reply to Thread

Guest, we'd love to know what you think about the forum! Take the Diabetes Forum Survey 2025 »

Message: <blockquote data-quote="KennyA" data-source="post: 2576539" data-attributes="member: 517579">Interesting question. I worked for years somewhere that had all the data you could possibly want on just about everything, but had actually very little information, because very few people actually bothered analysing it.  For example we used to report performance in a particular sector as an "average" (mean) of all the individual units, and this came out (iirc) at about 68%.  However once you picked it apart, it became clear that most units were scoring in the 90%s (the median was 93) but we had a troublesome group in the 20s and 30s, with nothing in between.  Nobody actually scored 68% and treating the high and low groups as if they did meant we got it wrong consistently. It was because that was how the spreadsheet had been set up originally, and it had never been questioned.It's achieving understanding that's crucial, and actively using data to drive the understanding. That should give you useful information.</blockquote>

Top Bottom