Jump to content
Great War Forum

Remembered Today:

1911 Census Demographics - Regular Battalions


Recommended Posts

Title has been changed to reflect the broadening scope of the thread.

 

I am experimenting with some demographic data and using the 4th Bn KRRC data as a test bed. The chart below shows the birth place of the Riflemen serving in the 4th Bn KRRC in India in 1911. It is the last Census before the war and almost certainly 99.9% of these men would have served during the Great War. 

 

I have a small spreadsheet linked to a chart with the map of the UK (The map will change to a better one when I can find it and include both parts of Ireland). The idea is that I can collate a Battalion's data and pour it into the framework, and it generates a distribution chart showing eactly where the men came from. I have only populated the chart with the locations with 2 or more men. There are about 300 more places to add which will show as tiny dots as they are places where only one 4th Bn man was born. I thought I would share it here and see if it was of any interest. 

 

We can do this exercise with every unit of the British Army in April 1911 (assuming we can find the Census returns). It might be interesting. I have a number of battalion's data already. It should provide an indication of just how few men were actually born within the recruiting district of the County. The KRRC of course were recruiting nationally, but despite this England accounted for 95% of the recruits. Note how small the Glasgow point is.  Not many Geordies eor Makums either.

 

Work in progress.

 

MG

4 KRRC Birthplace.jpg

Edited by Guest
999 to 99.9
Link to post
Share on other sites

If you did this for 1756, you'd find an even greater geographical spread ;)

Link to post
Share on other sites
  • Admin

I would be interested in knowing about an application in how you are able to link a birth place and/or enlistment place to a map like you have done - is this a software application freely available?

 

Regards

 

Russ

 

  

Link to post
Share on other sites

Having developed a few software business applications that use GIS (Geographic Information Systems) in my time, the main challenge is usually cleaning up the data so the locations can be accurately geo-referenced.

 

Even plotting using modern post code centroids can give mixed results - e.g. dots showing people as being born in the middle of the Carneddau!

 

I'd be interested in hearing how you've done that Martin.

 

Link to post
Share on other sites
1 hour ago, RussT said:

I would be interested in knowing about an application in how you are able to link a birth place and/or enlistment place to a map like you have done - is this a software application freely available?

 

Regards

 

Russ

 

  

 No. I built it myself. There are a few applications online but they all have limitations. It was the frustration that eventaually led me to build my own. Some hard yards in creating the eastings and northings of each location but it only has to be done once.

Edited by Guest
Link to post
Share on other sites
1 hour ago, MBrockway said:

Having developed a few software business applications that use GIS (Geographic Information Systems) in my time, the main challenge is usually cleaning up the data so the locations can be accurately geo-referenced.

 

Even plotting using modern post code centroids can give mixed results - e.g. dots showing people as being born in the middle of the Carneddau!

 

I'd be interested in hearing how you've done that Martin.

 

 

Cleaning up the data can only be done with the Mk I eyeball. "Penrith, Wales" has to be done by eye and hand. It takes a few minutes to clean a battalion's worth of Census data.

 

I experimented with a few websites, applications and data-sets. For my purposes if a man says he was born in Grimsville, the grid ref of the Town Hall or Town Square suffices. I used a website that provides an Ord Survey grid and one can navigate the map, find the exact location and right click. Up pops a box with lots of data which can then be cut-and-pasted into the spreadsheet. Essentially one simply builds a reference data base of 1,000 town and villages in GB. Then I drop in the data from the Census. Copy the master list, and eliminate duplicates to get the unique list of locations and use the COUNTIF function to populate the incidence/frequency data to find out how many came from Grimsville, Lancashire or Coalville, Durham. The outputs are already within the framework of the Chart Data-Series, and therefore automatically populates. It can be quickly adapted to counts per County, in fact I think I will create two parallel charts one with town and and village and the other with County centred on the principal County Town.

 

It is not that difficult and something I have been meaning to do for ages. Each iteration adds a few dozen more locations. After 10 of them the new additions become fewer and fewer still

 

It could also be adapted for CWGC data and SDGW data. The only leg works is copying the long list and eliminating duplicates and then running the COUNTIFS. It takes about ten minutes with each new set of data. If the data had more refined addresses, then there would be more legwork, however on the Military Census of 1911 soldiers only entered Towns or villages and not street names or addresses. 

 

There are databases one can download for a cast that provide a myriad of Geo Referencing data for a postcode or grid and convert either way, however the raw data usually says 'Grimsville' so the town hall will do just nicely. It is really about trying to show the geographic distribution across the country rather than nailing individuals to No. 13 Clog  Street, Grimsville, Lancashire. 

 

My aims are really to try and plot the origins, by battalion of the BEF in 1914, however the framework does have other potential applications. MG

Edited by Guest
Link to post
Share on other sites

QUOTE: The chart below shows the birth place of the Riflemen serving in the 4th Bn KRRC in India in 1911. It is the last Census before the war and almost certainly 99.9% of these men would have served during the Great War. 

 

Whoa! there!

 

So only one in a thousand in 1911 were well into  "12 years with the colours" terms, and therefore under no obligation, and also did not re-engage on the outbreak of war?

 

 

Link to post
Share on other sites
3 hours ago, Muerrisch said:

QUOTE: The chart below shows the birth place of the Riflemen serving in the 4th Bn KRRC in India in 1911. It is the last Census before the war and almost certainly 99.9% of these men would have served during the Great War. 

 

Whoa! there!

 

So only one in a thousand in 1911 were well into  "12 years with the colours" terms, and therefore under no obligation, and also did not re-engage on the outbreak of war?

 

 

Yeah - I let that one by as it was so obviously wrong!

 

Also I am amazed there were no 4/KRRC riflemen born in Ireland!

 

Still highly worthwhile though.

Link to post
Share on other sites
7 hours ago, Muerrisch said:

QUOTE: The chart below shows the birth place of the Riflemen serving in the 4th Bn KRRC in India in 1911. It is the last Census before the war and almost certainly 99.9% of these men would have served during the Great War. 

 

Whoa! there!

 

So only one in a thousand in 1911 were well into  "12 years with the colours" terms, and therefore under no obligation, and also did not re-engage on the outbreak of war?

 

 

 

 

Hmmm... Lets assume a man was approaching Time Expired in 1911. If he had joined aged 17 he would be aged 29 (calc: 17+12=29) in 1911. By 1916 when the Military Services Act came in he would be 34 and still liable to be called up. 

 

Working backwards from the Military Services Act 1916: anyone aged up to 41 was liable for conscription. Anyone who was 41 in 1916 would be 36 in 1911. Of the 927 men on the 4th Bn KRRC Census for 1911, Only 18 men are over the age of 36. meaning 909 would definitely be in the frame for Service either due to current obligations or caught by the Military Services Act. That represents 98% of the men on the Census roll. 

 

When we look at the remaining 18, five are Officers who definitely served in some capacity during the war, which gets us to 98.6%, leaving just 13 men Aged between 37 and 43 or 1.4% of the Battalion.

 

As you know, in late 1914 AO 295 allowed ex-servicemen to re-enlist. The age barrier was 45, later raised to 50; that relates to men aged 40 and 45 in 1911. Only the Bandmaster and the Sergeant Major were older than 40 in 1911). Under this AO some 91,411 ex soldiers rejoined the Line Infantry or the equivalent to 617 men per line infantry Battalion. As you know the KRRC represented 4 of those 148 battalions, meaning it is likely that around 2,470 time expired KRRC soldiers rejoined between 1914 and 1915. Importantly it allowed NCOs and WOs to retain former rank. I would have a very strong bet that aside from death and medical discharge all of those remaining 13 soldiers (incidentally mostly Senior NCOs) volunteered and served in some capacity during the war, probably in the Reserve battalions in administrative roles even if medically downgraded I suspect they would have been given desk jobs in the Depot or Reserve Battalion or in a host of administrative roles that the expanding Army needed to fill. In 1915 the Army was still desperately short of NCOs. 

 

The KRRC raised 19 more battalions during the War and the two Reserve Battalions expanded to double the size of a normal battalion. Some rough calculations would suggest they would need around 1,000 NCOs and I would be amazed if these 13 old and bold soldiers did not respond to the bugle call. If the other Regular Battalions of the KRRC had similar numbers, there would only be around 50 in the whole regiment in this category, or  roughly 3 per newly raised battalion. The only caveats are death and medical discharge. 

 

None of the 927 would be barred due to age simply because of the AO 295. The only two who would be questionable on that front were the Bandmaster and the Sergeant major, both over 45 in 1916. I would be truly astonished if these two had not served in some capacity. 

 

Any mistakes are mine. MG

 

Edit. In 1918 The Military Services Act raised the upper age limit to 51. Anyone who was 51 in 1918 would be 44 in 1911. Every single Officer and man on the 1911 Census for the 4th Bn KRRC was under the age of 44. As I said, aside from death or medical downgrade these men would have served in some capacity during the war. 

 

Edited by Guest
Link to post
Share on other sites
On 11/5/2016 at 03:45, MBrockway said:

Yeah - I let that one by as it was so obviously wrong!

 

Also I am amazed there were no 4/KRRC riflemen born in Ireland!

 

Still highly worthwhile though.

 

1. See my response to Grumpy

 

2. The OS grid ref finder does not work for Eire. I had to approximate Dublin (shown)

 

Of the 927:

Cananda...........2

England.........889

India..................6

Ireland.............18  (Dublin 5, Cork 5, Kilkenny, Meath, Monkstown, Tralee 1 each)

Scotland............6

South Africa......1

United States....1

Wales................2

Unknown...........1

 

When updated I will post again. There is a long tail of places where a single person came from....

 

MG

Edited by Guest
Link to post
Share on other sites

MG

 

Very interesting topic.

 

How about plotting 1911 Census information against the 14 Star Roll? 1st Bn ESR could be an obvious trial. However, after having a look at the 14 Star Roll and trying to marry it to the Census I got more hits for the 2nd Bn and Kingston upon Thames so back to the drawing board.

 

Bootneck

Edited by bootneck
Woolly Thinking!
Link to post
Share on other sites

by way of comparison, 4th Bn Worcesters, dominant in the midlands. The 189 men of the 1,033 named men rather dominates the data. It is semi-transparent in order to show those groups hailing from Wolverhampton, Worcester, Kidderminster, West Bromwich etc. 

4th Bn Worcs 1911 Census.jpg

4th Bn Worcestesr Closeup.JPG

Edited by Guest
Link to post
Share on other sites
9 hours ago, QGE said:

 

 

Hmmm... Lets assume a man was approaching Time Expired in 1911. If he had joined aged 17 he would be aged 29 (calc: 17+12=29) in 1911. By 1916 when the Military Services Act came in he would be 34 and still liable to be called up. 

 

Working backwards from the Military Services Act 1916: anyone aged up to 41 was liable for conscription. Anyone who was 41 in 1916 would be 36 in 1911. Of the 927 men on the 4th Bn KRRC Census for 1911, Only 18 men are over the age of 36. meaning 909 would definitely be in the frame for Service either due to current obligations or caught by the Military Services Act. That represents 98% of the men on the Census roll. 

 

When we look at the remaining 18, five are Officers who definitely served in some capacity during the war, which gets us to 98.6%, leaving just 13 men Aged between 37 and 43 or 1.4% of the Battalion.

 

As you know, in late 1914 AO 295 allowed ex-servicemen to re-enlist. The age barrier was 45, later raised to 50; that relates to men aged 40 and 45 in 1911. Only the Bandmaster and the Sergeant Major were older than 40 in 1911). Under this AO some 91,411 ex soldiers rejoined the Line Infantry or the equivalent to 617 men per line infantry Battalion. As you know the KRRC represented 4 of those 148 battalions, meaning it is likely that around 2,470 time expired KRRC soldiers rejoined between 1914 and 1915. Importantly it allowed NCOs and WOs to retain former rank. I would have a very strong bet that aside from death and medical discharge all of those remaining 13 soldiers (incidentally mostly Senior NCOs) volunteered and served in some capacity during the war, probably in the Reserve battalions in administrative roles even if medically downgraded I suspect they would have been given desk jobs in the Depot or Reserve Battalion or in a host of administrative roles that the expanding Army needed to fill. In 1915 the Army was still desperately short of NCOs. 

 

The KRRC raised 19 more battalions during the War and the two Reserve Battalions expanded to double the size of a normal battalion. Some rough calculations would suggest they would need around 1,000 NCOs and I would be amazed if these 13 old and bold soldiers did not respond to the bugle call. If the other Regular Battalions of the KRRC had similar numbers, there would only be around 50 in the whole regiment in this category, or  roughly 3 per newly raised battalion. The only caveats are death and medical discharge. 

 

None of the 927 would be barred due to age simply because of the AO 295. The only two who would be questionable on that front were the Bandmaster and the Sergeant major, both over 45 in 1916. I would be truly astonished if these two had not served in some capacity. 

 

Any mistakes are mine. MG

 

Edit. In 1918 The Military Services Act raised the upper age limit to 51. Anyone who was 51 in 1918 would be 44 in 1911. Every single Officer and man on the 1911 Census for the 4th Bn KRRC was under the age of 44. As I said, aside from death or medical downgrade these men would have served in some capacity during the war. 

 

 

Martin, you make a good case, I am almost convinced, although I suspect that mortality and illness might just defeat you.

 

Either way, well done, the cigar is yours!

 

PS. Don't fret over the Bandmaster, because bandmasters did not deploy.

Edited by Muerrisch
afterthought
Link to post
Share on other sites
10 minutes ago, Muerrisch said:

PS. Don't fret over the Bandmaster, because bandmasters did not deploy.

 

On a pedantic note, all of the bands of the Foot Guards regiments went overseas. The Grenadiers did so in time to get the 1914-15 Star and claim to be the only Band that qualified for the Star. The Guards took it in turns to go out. I suspect our Bandmaster however might well have been employed raising morale back in Blighty. 

 

I think the Sergeant Major may have ended up with the 4th Bn OBLI TF.

Link to post
Share on other sites

Very interesting stuff Martin.  It will be a useful project.

 

A glitch that needs resolving: On your zoomed in map of West Midlands, the dots seem to have mis-aligned ... or at least there's no dot on Kidderminster anyway!  And is the large blob about 20km E of Worcester supposed to be centred on Worcester?

 

Could you have worked off different map datums?

 

I immediately looked at Kidderminster because my grandfather lived there  ... but enlisted into the 60th, not the Worcs Regt!

 

Cheers,

Mark

 

 

Link to post
Share on other sites
3 hours ago, MBrockway said:

Very interesting stuff Martin.  It will be a useful project.

 

A glitch that needs resolving: On your zoomed in map of West Midlands, the dots seem to have mis-aligned ... or at least there's no dot on Kidderminster anyway!  And is the large blob about 20km E of Worcester supposed to be centred on Worcester?

 

Could you have worked off different map datums?

 

I immediately looked at Kidderminster because my grandfather lived there  ... but enlisted into the 60th, not the Worcs Regt!

 

Cheers,

Mark

 

 

 There is some shift in the X axis. The plots all use Ordnance Survey data and the map is an image that is and underlay to the scatter diagram. I have to adjust the minimum and maximum of the X axis scale to get the underlying map to 'fit' to the data. Ditto the Y axis. If I get the Midlands data to line up, London shifts towards Reading. 

 

I think the 'underlying' (sic) problem is that the OS coordinate and the map are based on quite different datum (as you spotted). Curvature of the earth etc. I have had this challenge before with Gallipoli map overlaid onto Google Earth; If I get one part of the map to align, the rest slips. One ends up going round in circles. There is the possibility that the map was slightly compresses longitudinally, so I might play around with stretching it and reloading. 

 

As mentioned at the beginning, this is merely an experiment and I will trial different maps. I will continue to use OS Eastings and Northings as they can be converted (if necessary) to other coordinates, depending on the type of map. It is a 95% solution at present, and given it was all done on Excel, not a bad first attempt. The relative positions of the dots and their relative size are correct as they are all based on hard data. The small challenge is the underlying image of the map. MG

 

Work in progress

Edited by Guest
Link to post
Share on other sites

I've had similar issues with plotting the Post Code Sector parts of Mountain Rescue search managers' home addresses onto a Google Map, so our proximity to major incidents (floods, aircrashes etc.), & key details, are rapidly visible to the Police and other agencies.  Part of our Local Resilience planning.

 

It was directly due to map datum and projection differences.  Plus some minor errors from the freeware Post Code geocoding tools I was using.

 

I got round it temporarily by manually altering the eastings and northings of the derived OS grid refs for the PC Sector centroid in the spreadsheet that drives the map display.

 

Since I was only displaying the centroid of the Post Code Sector, NOT each person's actual home address (Data Protection issues), I could get away with this.  Also I only had a few dozen names to handle.

 

Eventually I'll use a better map background, but the Google Map route was quick, cheap and cheerful!

 

I'll have a think about your project and how you could solve this issue.

 

Cheers,

Mark

 

Link to post
Share on other sites

Encouraged by my early results with excel charts, I have just started to grapple with QGIS which seems to be everyone's preferred option for free software for digital mapping. It looks quite sophisticated for anyone like me with no mapping/surveying knowledge. Separately the Ordnance Survey has a list of maps that one can download for free in digital format with various layers. These can also be incorporated. I have go as far as downloading a map of the British Isles and uploading it to QGIS. As one passes the cursor over the map one sees the Grid references. It goes do down in incredibly small scale, so if one wanted to plot addresses this is possible. Also the OS has a free download of every town and village running to tens of thousands of locations with;

 

Grid Reference

Eastings

Northings

Latitude

Longitude

Postcode

 

It is a major step forward as I have, to date, been entering the Eastings and Northings manually to create my own database. I have set up a simple function in excel that searches the data for a matching word such as 'Kettering' and then automatically populates the Eastings and Northings data

 

Thinking aloud, it would be easy to set up the spreadsheet with places running top to bottom in column A, then Columns B to G with the data above and then have separate columns for each unit. With the Census data one could simply drop each unit's data into a column and then there would be a Master database of the British Army by birthplace in 1911.One could then simply select which unit or units one wanted to see on the bubble chart. Neat. 

 

I already have a few Battalion's worth of data based on the 1911 Census of military units.The weakness of course is that it will not capture the Reservists, however it will at least capture a very large per cent of the regulars, particularly those overseas in India etc where units were at full strength, and As pointed out earlier, the vast majority would have served during the war due to existing commitments under terms of engagement, or if time expired by 1914, probably caught agin by the Military Services Acts of 1916 and 1918.

 

I think it would be fascinating to see the demographic spread of each unit. If one could find data on TF units or Kitchener units, again it would be interesting to see the diffusion of the men's place of birth. It would be interesting to compare, say, Scottish battalions or battalions within a Regiment. I strongly suspect there will be a heavy tilt towards the metropolitan areas.

 

Separately the SDGW and CWGC databases might provide statistically meaningful samples of units and their men's places of birth, enlistment and residence. Again tis could be quite revealing for, say Kitchener Battalions recruited from a particular geographic area. Ditto the TF where we dont have the 1911 Census data by unit. Lots of potential here. 

 

I suspect it will take a while to get to grips with QGIS.

 

For my purposes this becomes really quite interesting. I am focused on linking transcribed diaries with medal rolls, particularly the 1914 and 1914-15 Star rolls; linking two separate databases is providing some unique insights as I am able to track when men arrived and (occasionally) departed. It also puts names to the thousands of men who served and eactly when they served. If I can then link these names to demographic data such as birth place, residence and place of enlistment, one begins to add a human dimension. The potential to link thousands of individuals by origin to a particular unit and its subsequent history will be a unique perspective. 

 

I am thinking of databases such as the Scottish National War Memorial data. Most of the 132,000 odd names have the necessary data. It would be interesting to see these plotted on a map, particularly all the Englishmen serving in Scottish units.  Lots to think about. 

 

MG

Link to post
Share on other sites
18 hours ago, QGE said:

 

On a pedantic note, all of the bands of the Foot Guards regiments went overseas. The Grenadiers did so in time to get the 1914-15 Star and claim to be the only Band that qualified for the Star. The Guards took it in turns to go out. I suspect our Bandmaster however might well have been employed raising morale back in Blighty. 

 

I think the Sergeant Major may have ended up with the 4th Bn OBLI TF.

 

Perhaps not all readers will be aware that Foot Guards regiments had regimental bands as opposed to a band for each battalion. Again, unlike the Foot, the regimental band did not have a formal established war role as, for example, stretcher bearers. Neither were Foot Guards bands "supposed" to deploy on active service. Each Guards battalion relied on its Corps of Drums [drum, flute and bugle, with pipes also for the Scots and perhaps Irish] for organic music.

 

Where there's a will there's a way, and good for them, because military bands have an important morale role.

Link to post
Share on other sites

QGIS is a free piece of software that I have spent all morning trialling. The first attempt at doing a scatter diagram of the 1911 Census birthplaces of the members of the 4th Bn Worcestershire Regt are below. One dot for each recorded location. What I have not yet cracked is making each dote larger if there is more than one person from theat location. It is however a huge step forward as it uses longitude and latitude down to five decimal places. No more X Axis slippage.

 

The undelying map is a bit rubbish but I am working on it. 

 

Separately I have manage to buy a database of 48,000 towns and villages with longitude, latitude, OS Grid ref, eastings and Northings and Postcodes, region, county etc. All for £30. A bargain. I can now use this as the reference database to populate the battalion sheets. A simple list of birthplaces and the numbers  can be populated with the data in less than a few seconds. Now that I have the framework in place it should be markedly easier. 

When we zoom in we can see the granular distribution much easier, and you will note they are fixed to the correct location. Simply marvellous for a bit of sophisticated free software. It could be useful for War Memorial work, plotting the known homes etc. I will tray and do some SNWM sampling and colour code Scots born and English born etc. It could be interesting.  Mg

QGIS Trial 1.JPG

Link to post
Share on other sites

Definitely a big improvement.  Google maps is OK so far as it goes, but it is flawed when you try and use it for more sophisticated stuff.  I also find it very slow to refresh/reload on moving around, especially since the last upgrade.  Much better to use better open source maps and GIS API's ... and without all the sinister tracking that Google is prone to!

 

I'm sure you'll find ways of embedding QGIS charts into webpages that are just as easy as creating a new Google Maps map.

 

One caveat about the gazetteer tools - be careful about simple word matching - your 'Kettering' could get hits other than the Kettering, Northants., that you expect and it's very difficult to spot a dot that's wrong on the map in a large dataset.  Are you able to select one of the dots/dot clusters and 'drill down' to see the source data?

 

How about trapping it with code in your Excel for detecting multiple matches?

 

You need to get it right, because such powerful maps will quickly get propagated across cyberspace and future researchers will use them as cast iron sources!

 

Mark

 

Link to post
Share on other sites

Incidentally IIRC there is good Gazetteer data from the OS available free.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...