20 years of books and cakes: a book group’s data story

My book group is 20 years old this year. Something that I’m pretty sure none of us imagined would happen back at the very first meeting in autumn 2002.

Since then we’ve had at least 18 members, read at least 167 books and eaten at least 110 cakes.

Sadly, it didn’t occur to us to keep a record of what we were reading until I tried to reconstruct it from memory in March 2012. It’s definitely incomplete and the dates are often guesstimates. From then, I kept a note of the books, but it wasn’t several years later that I started to systematically record other details: the person who had made the selection, the date and the cake we ate with the book. 

But with all those caveats, here’s what our (slightly flaky) records reveal.

Who are the readers?

Here’s the members and the percentage of choices for each person.

Not everyone has been a member for 20 years. A few were with us for just a few years or less, some have left altogether, others came in later, and some have come and gone, and then come back again.

Of the long-standing and most consistently active members you can see that the selections are fairly evenly divided. Between Jackie, Astrid, Maryanne, Lucy, Emma, Val and Morag, the choices range from 7.2% to 13.8% of the overall selections.

Though for 9.6% of the books we can’t remember who chose them!

What are we reading

Here’s the genres we’ve read.

This is really just based on my identification of a work as belonging to a single genre so there may be some points of dispute. I designated The Princess Bride as satire, not fairy tale. And is Orlando best described as fantasy? The Truth is also tagged as fantasy though it could also be categorised as comic or satire. And is there really a difference between historical fiction and historical crime?

Anyway, ignoring these knotty issues you can see that most of our consumption is general fiction (40.7%) and historical fiction (21%). That was as expected however I was a little surprised that memoir came in at almost 10%.

Crime was at 6% though there are also two thrillers (The Da Vinci Code and Tinker, Tailor, Soldier, Spy) and two historical crime fiction (The Observations and When We Were Orphans) plus one true crime which was In Cold Blood.

Sci-fi is at 6% and fantasy at 3.6%. We don’t really read full-on horror though we have tried one gothic novel– that was We Have Always Lived In The Castle by Shirley Jackson though perhaps Wide Sargasso Sea and Frankenstein would also qualify as gothic – and four that I have classed as supernatural (2.4%).

Our one book of fairy tales was Oscar Wilde’s The Happy Prince and Other Tales. There have been five comic novels (3%) and four satirical works (2.4%). The solitary erotica (0.6%) was Fifty Shades of Grey (of course!).

Finally, the self-help book was a slight anomaly as we took self-help as a topic and we each picked whatever took our fancy. I chose How To Make Friends and Influence People which is the one recorded in the data. Though we did also read Untamed which I’ve classed as memoir but I guess there is an argument that it strays into the self-help arena.

I wondered if certain readers preferred particular genres. Again, sometime I only have a few choices recorded for one person, so it’s not necesarily compelling evidence for a preference.

Val has the most variety, notching up 10 genres on her bookcase, followed by  Morag, Lucy and Astrid who have each selected from eight genres.

Jackie and Astrid have led on memoir with four and three selections respectively, followed by Emma, Lucy and Claire having each selected two.

For historical fiction, Morag is top (eight choices), then Emma (six), and Val (five). Lucy and Josie (Sa) both have two fantasy selections, while Val and Emma have picked one each.

Maryanne has the most sci-fi (three choices), followed by two for Emma, and one each for Morag, Lucy, Astrid, Jackie and Bev. It’s also worth noting that – although it’s not recorded as a genre because I think it’s a format – sci-fi also accounts for the only graphic novel we’ve read; that was Watchmen, as chosen by Jackie.

From where are we reading?

I looked at where the books were originally published to see what kind of geographical spread we’ve gone for.

Fairly predictably, the UK is top with 96, followed by the USA with 45. Then there are five from Canada, four from Australia, two from France, and one from all of these countries:

We do read only in English. Of the 9% of the total books that we have read in translation the original languages of publication are: French (five books), Russian (three books) and one book each in Japanese, German, Italian, Korean, Norwegian, Swedish and Turkish.

What about Scottish books?

We’re based in Scotland so you’d think that would have an impact on what we read. So do we favour Scottish books?

I defined Scottish books as those with these connections:

1. The author was born in Scotland, is of Scottish descent, or they were living/working in Scotland when the book was produced, for example Arthur Conan Doyle, Muriel Spark, Louise Welsh and Maggie O’Farrell. 
2. The book has a Scottish setting or deals with the work or life of a Scot or with a Scottish question, event or situation.

So, yes, 18% of our books have a ‘Scottish’ author as defined above. Which is quite a bit chunk of all our reads.

There’s also one book (0.6%) where the author isn’t Scottish but it is set in Scotland. Which book? Outlander, obviously!

From when are we reading?

Do we favour books from particular periods? The oldest book we’ve read dates from 1759 (Candide), and we have a smattering from the 19^th century:

• Frankenstein (1818)
• Dead Souls (1842)
• Romola (1863)
• The Happy Prince and Other Tales (1888)

But the vast majority were first published in 20^th and 21^st centuries. This is how they are distributed by decades.

You can see a significant leaning towards books that have been published around about the same time that book group has been going – the 2000s (58) and 2010s (53) and already 10 from the 2020s.

Combining the dates of publication with the dates when we read the book (data is patchy!) means we can calculate an average length of time between original publication and reading. At this point in time this Zeitgeist Gauge stands at 18.8 years!

Who are we reading?

The authors are 60.5% female and 39.5% male. We haven’t (knowingly) read any authors who are non-binary. And there are more female than male authors in almost all genres apart from:

• satire (three men: one woman)
• thriller (two men and no women)
• historical crime fiction (one man and one woman)
• and true crime, self-help and fairy tale, each of which are represented by one male author (Truman Copote, Dale Carnegie and Oscar Wilde).

I’m the only person with more than two selections who has picked more male than female authors; out of 18 selections, 10 are men. Most people have more women than men, except Claire (four selections in total) and Jane (three selections) who have only picked women writers.

What are we eating?

I like baking, so for the very first meeting in 2002 I brought a homemade, red wine and chocolate cake. It seemed like a good combination – books, wine and chocolate.

At some point, when we read Frankenstein, possibly in 2003 or 2004, we had an epiphany: to match the cakes to the books! I’ll probably write more about that process in a later blog post. And I should point out that it’s not the only factor in choosing a cake – other factors such a time of year and seasonality often also play into the decision.

But to find out what we’ve been eating I categorised the cakes by type. Here’s how it turned out.

By far and away the most predominant type are sponge cakes at almost 60%. No surprise perhaps; they’re quick and easy to make and can carry many different flavours.

Removing the 27% of meetings where there was either no cake or it’s not been recorded, then that leaves 13% for all the others. This is how they ranked in terms of popularity.

Type	Percentage (%)	Total number
Fruit cake	6	10
Pie	4.8	8
Cheesecake	3.6	6
Tart	3.6	6
Pastry	1.8	3
Biscuit	1.8	3
Sweet bread	0.6	1

For information, the pastries were:

• baklava for My Name is Red
• Gateaux St Honore for Suite Francaise
• croque em bouche for The Princess Bride

The biscuits were:

• gingerbread house for The Happy Prince and Other Tales
• Empire (German) biscuits for All Quiet on the Western Front
• Laura Bush’s Texas cowboy cookies for American Wife

And the sweet bread was an Italian Easter bread which was an emergency purchase for The Master and Margarita.  For the record I’m pretty sure I’ve made a gugelhupf (which has yeast in so would also be classed as a sweet bread) as but I can’t remember for which book that was.

So are certain cakes were more associated with particular genres?

It’s hard to see a clear association but perhaps that‘s because of the dominance of sponge cakes in the data. The types seem to be pretty well spread across all genres.

Memoir was the one exception. Of the 16 books we read in this genre, the cakes for five are not recorded and two had no cake. Of the remaining nine, there were four sponge cakes but, in contrast to the other genres where sponge is frequently the most numerous type, in this case, sponge cake and fruit cake are equal with four appearances each. I suppose it’s understandable that books reflecting on the past lend themselves to being accompanied by a fairly traditional cake. (The remaining cake in this genre was a tart).

Though perhaps then its curious that historical fiction has only one fruit cake associated with it (Molly cake for The Fatal Tree).

I also tried to categorise by flavour but that’s quite challenging as cakes may have several flavourings or no specific flavour – ie they are just plain sponge cake but may have toppings or fillings of fresh fruit, cream,  jam etc.

However you can see that the dominant ones are chocolate , vanilla and dried fruit.

One that may be a bit surprising is caraway – it’s a personal fondness of mine – though I was intrigued to find that it seems I’ve used it entirely for books published in the 1930s (Gaudy Night, Cold Comfort Farm) or with strong connections to that decade (Larchfield).

What more could be learnt?

There’s probably a lot more that could be done with the data.

We could do lots more analysis of the books, such as:

• The number with male or female lead protagonists.
• The division between ‘happy’ or ‘sad’ endings.
• The locations in which they are set.
• The eras in which they are set.
• Whether they are third person, first person, or second person narration.
• The ethnicity of the authors – including analysis of whether our choices have got more diverse over the years.
• The type/size of publisher: Big Five, medium-sized, small independent, self-published.

Or we could compare our book choices with public book group selections like Richard and Judy, or see how many of our reads have won literary awards.

For the cakes I guess the caraway connection mentioned above suggests there’s the capacity to dig a bit deeper into whether certain flavours are consistently associated with particular eras or settings, as well as genres.

The full dataset is in Google docs if you want to see more of the details such as the titles of all the books. It’s a living document, so I’ll continue to update it and that in turn will update the live data report where these interactive charts and graphs are published. So this blog post is a snapshot in time to mark our 20th anniversary.

This is a personal archive for me, though I wish I’d kept the records in more detail and in better order from the start – a lesson to learn there!

Assuming that book group continues and more data is added perhaps some patterns will become clearer or new patterns will emerge. Though I suppose that now I have more insights into the connections between books and bakes it’s feasible that the knowledge may influence my future cake choices.

Which means I need to ask, could this data be used to predict a cake choice?! I do hope not; it’s nice to have some plot twists that can’t be foreseen!