363 English pound cake recipes. For every word used in at least 10 of their titles, the recipes that use it against the rest, on every ingredient, in grams per 100 g of flour. 2 words go with a clear difference; 7 others change nothing measurable.
This is what recipe writers do when they pick a word, not proof that an ingredient makes the texture: the numbers show what recipes with the word have in common, and a recipe can earn its word through the method, the pan or the oven instead.
Words that name something in the recipe or the kind it is. Some differences here are what the word says outright; they show the numbers can be trusted where the answer is known.
Recipes with “orange” in the title: less butter (61% vs 91%); more leavening (2.5% vs 1.3%); less sugar (107% vs 133%).
“orange” (15)the rest (348)middle half of all
Butterrecipes with orange in the title 61% vs the rest 91%
Leaveningrecipes with orange in the title 2.5% vs the rest 1.3%
Sugarrecipes with orange in the title 107% vs the rest 133%
For example: Orange Pound Cake (crazyforcrust.com) · Orange Pound Cake (momontimeout.com)
Recipes with “chocolate” in the title: more dairy (78% vs 58%).
“chocolate” (19)the rest (344)middle half of all
Dairyrecipes with chocolate in the title 78% vs the rest 58%
For example: Chocolate Pound Cake (beyondfrosting.com) · Chocolate Pound Cake (foodnetwork.com)
Recipes with these words in the title use about the same amounts as the rest on every ingredient, within what chance would explain. The number is how many titles use the word.
Titles are split into words, with two-word names kept whole ("small batch", "cream cheese", "brown butter") and spellings folded together ("crisp" and "crispy"). The dish's own name, words like "the" and "recipe", and any word in half the titles or more are left out. A word needs at least 10 recipes, and there must be 10 without it. English titles only, so a word cannot stand in for a language.
A difference is shown when the medians sit at least a third of the middle half of all these recipes apart (about half a standard deviation, a medium effect) and a rank test puts it at 3 standard errors or more. For an ingredient many recipes skip, the share using none counts too: 25 points apart at the same 3 standard errors. With dozens of words and every ingredient tested, about one chance difference in 370 comparisons would pass, so a lab may show one that is luck; a finding that repeats across labs is unlikely to be.
Shares are rounded down. Each card links two recipes that use the word, nearest its own medians.