Available online
Description
Counting things : How hard can it be to count things? You would be surprised. More or Less investigates. Walk into a field and count the sheep. There are three. But one is a lamb. So maybe we shoul...d say 2.5. And one of them is a pregnant ewe. So maybe we should say 2.75. If something so simple gets tricky so quickly, imagine the problem of counting millions of visits to hospital. Then try measuring how successful each one is, comparing every surgeon with every other, collecting all the data from people already under pressure - some of whom are not inclined to cooperate - making sure the data is comparable. Finally it would have to be summarised in a form that can be presented to a patient in a GP's consulting room. This massive task is arguably what is implied by Patient Choice, the government's new policy for the health service in which patients are given a choice of four hospitals for their treatment, but within two years will, the government says, be able to choose throughout the health care system. Data capture : In More or Less this week, we investigate the problems of what statisticians call data capture. If the typical impression of gathering information is that you go out, write it down and key it into a computer so that it is there for all to see on a readily accessible database, that impression could not be more wrong, says one of our contributors this week. Find out how serious underperformance has been missed, what information we really do have to inform choice, and the effort required to improve data for patients by listening to this week's programme. Judge for yourself what the simple problem of counting things means for the government's policy ambitions.
Counting things : How hard can it be to count things? You would be surprised. More or Less investigates. Walk into a field and count the sheep. There are three. But one is a lamb. So maybe we shoul...d say 2.5. And one of them is a pregnant ewe. So maybe we should say 2.75. If something so simple gets tricky so quickly, imagine the problem of counting millions of visits to hospital. Then try measuring how successful each one is, comparing every surgeon with every other, collecting all the data from people already under pressure - some of whom are not inclined to cooperate - making sure the data is comparable. Finally it would have to be summarised in a form that can be presented to a patient in a GP's consulting room. This massive task is arguably what is implied by Patient Choice, the government's new policy for the health service in which patients are given a choice of four hospitals for their treatment, but within two years will, the government says, be able to choose throughout the health care system. Data capture : In More or Less this week, we investigate the problems of what statisticians call data capture. If the typical impression of gathering information is that you go out, write it down and key it into a computer so that it is there for all to see on a readily accessible database, that impression could not be more wrong, says one of our contributors this week. Find out how serious underperformance has been missed, what information we really do have to inform choice, and the effort required to improve data for patients by listening to this week's programme. Judge for yourself what the simple problem of counting things means for the government's policy ambitions.
| Series: | More or less |
|---|---|
| Episode | 3 |
| First transmission date: | 2006-01-05 |
| Published: | 2006 |
| Rights Statement: | Rights owned or controlled by The Open University |
| Restrictions on use: | This material can be used in accordance with The Open University conditions of use. A link to the conditions can be found at the bottom of all OU Digital Archive web pages. |
| Duration: | 00:26:53 |
| + Show more... | |
| Producer: | Michael Blastland |
| Contributor: | A. W Dilnot |
| Publisher: | BBC Open University |
| Subject terms: | Mathematical statistics; Social sciences--Statistical methods |
| Production number: | AUDA040B |
| Available to public: | no |
