Monday, 29 December 2014

This Year's Christmas Cards in Statistics (or... "How bored can one person be at Christmas?")

Someone said to me the other day "it's a very glittery year for Christmas Cards, isn't it?".  So that got me thinking - how many Christmas Cards do have glitter on them these days?  And so I looked through all the cards sitting on our sideboard, and I counted.  The result is below...


So there you have it.  40% of the cards we received this year have glitter on them.

But that's not very rigorous, is it?  So before I start going deeper into some stats about this year's Christmas cards, let me tighten things up a little bit.

Firstly, sample size.  Well to tell you how many Christmas cards we received would be vulgar, and definitely "not a very British thing to do" so I'm not going to tell you the sample size.  But I will tell you that it's large enough that these figures are pretty meaningful.  Of course, the sample is biased by the fact that we only received cards from people we know - and we probably do know people who fit more tightly into some demographics than others.  I've made no attempt to correct the results based on the demographic of the people sending the cards, except in one particular case which I'll come to later.  I can also tell you that I've excluded corporate cards, and those received from companies and organisations.  

Secondly - what do I mean by "glitter"?  

Well I've taken "glitter" to mean that the front of the card has some "glittery substance" on it.  "Shiny" isn't enough.  There has to be something sparkly and metallic sprinkled on the front of the card for it to count.

Of course, this doesn't tell us whether this year is "more glittery than last year" as this only looks at this year's cards.  But still, it did get me thinking.  What else can be tell by looking at this year's Christmas cards?

Well let's stick with glitter for the moment.  "Snow" is something which is also featured heavily on Christmas cards.  And "snow" can also be "glittery".  But you can also have "glitter" without "snow" or indeed "snow" without "glitter".  Or you could have neither.  So here's the Venn diagram of the "glitter" and "snow" on the cards this year...


Again - let me tighten things up by saying that I've taken "snow" to mean that there is some representation of snow on the front of the card.  The ground being coloured in white or there being a single snowflake drawn on the cover counts. So it seems you're much more likely to find a card with "snow but no glitter" than a card with "glitter and no snow". Fascinating stuff, I'm sure you'll agree. 

So how about charity cards?  If you see a card sitting on someone's sideboard in their house, you can instantly see whether it's glittery or not so can you use that to make any judgement about whether it's likely to be a charity card?  Let's see...


 Well the answer is "not really" but "maybe slightly".  If the card is glittery then it's slightly less likely to be a charity card than it if doesn't have any glitter.  But it's marginal, so I wouldn't use to judge the charitable status of your friends.  (Well actually, I guess this would be checking the charitable giving of your friend's friends wouldn't it?)

I can hear you asking "but you've told us what percentage of cards are charity cards" or "what percentage of cards have snow on them" at this point.  Au contraire, my friends.  I haven't told it to you explicitly, but you can certainly work it out from the three graphs above, if the mood takes you.

So that covers glitter on cards, but then I started to think about something which someone else said to me.  "I think women are more likely to write the Christmas cards out than men!".  So let's look at whether the cards were written out by a man or a woman.
So it appears that yes - the majority of cards (72% in fact) are written by women.  I should again say that I judged this by my knowledge of handwriting in the cards, and using other methods (such as looking at whose name is written first, and in some cases asking people...) to have a very good guess at whether the card was written by a man or a woman.  I'm pretty confident in my assessment here.

And for the sake of rigour, I should say that if different people have signed their names on the card, I have taken the card to be written by the person who wrote the names of the recipients.  Of course, in these cases, by matching the handwriting of the greeting to the various samples of handwriting in the senders names, you have another way to find out who wrote the card...

"But no!" I can hear you cry.  "What about those women who live alone and so are bound to have written the cards?"  And you're absolutely right.  The above simply isn't fair.  In one-person households, or gay relationships, there isn't much option about the gender of the people who can write the card.  So I went through the statistics again.  This time removing all of the gay couples, and the people who live alone or whose circumstances otherwise mean that there wasn't a real choice about the gender of the person who would write out the card.  Of course, this also removed some men from the list as well as some women.  What happens now, then..?
As you can see, if you adjust the figures, then 80% of the cards were written by women - and the sample size is still large enough to be significant, in case you're wondering.  So I'm going to stick my neck out here, and say that yes - it's true that - well amongst people who send us cards at least - women tend to write more of the cards than men - by quite some margin!

So now that we're looking at the inside of the Christmas card, what's printed inside?  

I've made some linguistic judgements on this one.  I've distilled what's written in the card, and taken the "main greeting".  So something which says "Wishing you a Merry Christmas" gets lumped in with "Merry Christmas" and something which says "Greetings Of The Season" gets included with "Season's Greetings".  You get the idea.

So it seems that even in these days of "Winterval" and "Holiday Celebrations" people still tend to send cards which say "Merry Christmas" inside them.  

But how about what people write inside the card?  Besides the printed greeting, just how many words do people tend to write.  The minimum possible is - realistically - just the name of one person.  But it's more usual to write who the card is to (although we have 3 cards for which that isn't true...) and so given there are two of us that's two, three (if there's an "and" or "to") or four (if there's both) before you've started to write personal greetings and the name(s) of who sent the card.  Of course, some people tend to get slightly carried away, and write a letter inside the card.  I should say that I've only included words written inside the card itself.  If someone has included a Christmas "family newsletter" then I've not included that in the count. So just how much do people write in cards..?

I guess the answer is "not very much besides who the card is to and from".  I guess that's not surprising given it's Christmas, and people are writing a lot of cards. I wonder whether people tend to write more in Birthday cards...?

And those two outliers.  Some people do write a lot in cards, don't they?

So, now that we know what tends to be printed in the cards, and how many words of their own people tend to add - is it just Christmas, or do people talk about New Year, too?  Below is a graph showing how many cards mention New Year greetings as well as Christmas greetings...
I find this one quite interesting.  It seems that the vast majority of people don't wish you a "Happy New Year" in Christmas cards.  That surprised me...

So that's what people are writing, but how are they writing it?  In case you were wondering, here's what colour ink pepole are using to write the cards...
And yes - that means some people do write out their Christmas cards in red ink.

So who are these people who are sending the cards?  Friends, family, neighbours?
Now let me talk a little bit about this one, as I can hear you shouting at the screen "but I have some people who are both friends and neighbours!".  Well that's true.  But I've listed only the primary way in which we know someone.  So if they live in the same apartment block as us, they are listed as a "neighbour" even though some of the people in this category are most definitely "friends" and good ones at that.  And the same goes for family, too.  If someone is related to either one of use then they are counted as "family".  And if someone isn't either a "neighbour" or "family" under these definitions, then they are counted as a "friend".

So, I hear you shouting, what on earth is the point of all this?  I've already shown that you can't use a quick glance at a card and use whether it's "glittery" or not to work out whether it's a charity card.  But if you find yourself in our house around Christmas, there is something you can tell about a card by giving it a quick glance.  Have a quick look inside the card and count the words - not need to be exact - just roughly count the number of hand-written words...  then look at the graph below...
The graph above shows - for any given number of written words in a card - who is the card likely to be from.  And there's a clear pattern here - the more words written in the card, the more likely it is to be from someone who is neither family nor a neighbour.

There you go - who said statistics were useless...?

No comments:

Post a Comment