The National Institute of Health Research (NIHR) publish an annual table of national research activity. In this demo, I am going to use this dataset in Google sheets, by extracting a subset of the information from the website with a formula, and then using a few different pivot formula to manipulate the data.

The table is published on the webpage here

### Importhtml

To extract just the information for NHS organisations in Kent, Surrey and Sussex, in my Google sheet I have used the formula=query(importhtml("http://www.nihr.ac.uk/research-and-impact/nhs-research-performance/league-tables/league-table-data.htm", "table",1),"SELECT Col1, Col2, Col3,Col4,Col5,Col6,Col7,Col8 where Col3 ='Kent, Surrey and Sussex'")

### Simple pivot

To pivot the results, you can use a pivot table in Google sheets, via the data menu, However you can also use a formula to generate a pivotHere's a simple example (where the raw data is on a sheet called "Raw Data 1" - showing the number of recruits by organisation type.

=query('Raw Data 1'!A:H, "select sum(H) pivot B")

### Transpose a pivot

To change the orientation, so that it looks more like a normal pivot table, you can use 'transpose' in the formula=transpose(query('Raw Data 1'!A:H, "select sum(H) pivot B"))

### Filter a pivot

You can also apply a filter parameter in your query, so for example the following shows the sum of recruitment for just Acute trusts=transpose(query('Raw Data 1'!A:H, "select sum(H) where B ='Acute' pivot A "))

### Use multiple columns in a pivot

You can also bring other columns into a pivot, as long as these are included in the 'group by' or wrapped in an aggregate function. I have used a 'where' clause too to prevent blank rows being included in the pivot=query('Raw Data 1'!A:H, "select A,sum(H) where G >0 group by A pivot B")

### Sorting a pivot

Sorting a pivot is more tricky, in the example below, I have used a query, to query the query ;-)=QUERY( TRANSPOSE((query('Raw Data 1'!A:H, "select sum(H) where B ='Acute' pivot A"))) , "Select Col1,Col2 Order by Col2 desc ")

## No comments:

## Post a Comment