Remember last year? Remember the above figure (direct link), where all was laid out? This is how SCIENCE ranks your Halloween haul. Kit Kat and actual cash at the top, dental floss and anything whole wheat near the bottom.

Well, now it's time for the 2017 Candy Ranking Games. Go and fill out this 10 minute survey into the best and worst Halloween candies. Do it for science. Do it to fight #fakenews. Go forth, go go!

Note that data will be collected for analysis until noon, PST, Oct 25th.  This year’s Candy Hierarchy will be published on October 27th.

THINGS TO CONSIDER

You probably thought all we’d talk about this year is David S. Pumpkins and candy corn. Ask David S. Pumpkins to guest co-author, they said. He’s got a Wikipedia page and you don’t, they said. Just because that’s a year-old and only half-interesting reference, still, it’ll never get old and it’ll never die, they said, and by then they’ll forget you said something about candy corn in the first sentence. Because what the hell is going on with candy corn debates this year? How did 2017, of all years, become the one where candy corn blew up? It’s not a bell curve, it’s bi-modal, you know. People love it or hate it and never the twain shall meet. No twain meeting. You do know that don’t you? This comes up every year, obviously, given its importance, but wow, it’s all over the place this time. And then some dude told us they’re making Hershey bars with candy corn in them? What the dip-shit is that?

We don’t care, that’s what the dip-shit it is. We leave it up to you. That’s where we were headed with this. It’s up to you. Because David Ng and B.R. Cohen (that’s us) are again re-presenting the official universal survey about your candy favorites for The 2017 Candy Hierarchy.

If you want to spend your BoingBoing time reading about our vaunted methodology and sharp insights about survey logistics, if you feel like you need more debriefing on our statistical acumen and scientific ambitions, if you like talking about longitudinal studies and you’re not thinking of maps, go on ahead, read the back catalog. You’ll find that in just the past two years, we’ve tabulated 6000+ individual responses with over 600,000 preferences. But maybe you don’t care for numbers. Maybe you care only for the enormous skill the industry has for congealing sugar into bar, disc, dot, glob, cluster, dud, chip, wafer, cup, jack, cap, egg, or Donald Petersen. If you’re just here to define a candy hierarchy, that is, then have it. We give you the Fourth Annual Candy Hierarchy Survey. Go forth. Rank. We’ll be back next week to present this year’s hierarchy in full.

Click to view the below full-size; or download as a a high-quality digital poster (4MB) for detailed scrutiny; or proceed for the plain text, abstract and analysis…

"What’s going on with Kit Kats Dave?"

TRANSCRIPTION OF THIS MORNING'S CONFERENCE PROCEEDING DISCUSSION, WITH COHEN AND NG.

BC: What’s going on with Kit Kats Dave?

DN: I was about to ask you the same.

BC: Something’s going on with Kit-Kats.

DN: But what?

BC: That’s what I asked you.

DN: Something weird, that’s all I know.

BC: Because we have to start accepting a consensus result. Not counting the full-sized candy bars or hard cash—which are gimmes, we don’t even need to ask that—the year-after-year consensus has a pretty stable top 4.

DN: Kit Kat, Peanut Butter Cups, Twix, Snickers.

BC: Huge news there—Kit Kats put Peanut Butter cups in their place, kicking them down a notch.

DN: I’m sure that pleases you. So we can talk your peanut butter thing now.

BC: My Big Peanut Butter thing. I see two problems with Big PB, neither of them acceptable to me.

DN: You haven’t shut up about this for about five years. You’re about to go into your Mint ra—

BC: CHOCOLATE-MINT COMBOS ARE SUPERIOR TO ALL OTHER CHOCOLATE COMBOS, PB included.

DN: Thank you for screaming. And notice there is not one choc-mint combo in the top 20.

BC: I wasn’t listening, what?

DN:

BC: Doesn’t matter. I have other concerns. Like allergies.

DN: Medical science. You’ re trying to get us legitimacy?

BC: Yeah. We’ve come up to speed in most public eating forums on peanut allergies. But not Halloween. What gives?

DN: I have no reply to that.

BC: What’s the other big news this year?

DN: Yeah, let’s pivot.

BC: We have some good health news. People prefer “whole wheat anything” to “white bread.”

DN: Maybe. But people would also prefer Person of Interest Season 3 Box Set to a Box of Raisins.

BC: It’s not even their best season.

DN: You’re preaching to the choir.

BC: Bonkers the Board Game is preferable to Bonkers the candy.

DN: Most Just Born brand candies are mid-tier—

BC: Your Mike and Ikes, your Hot Tamales, your Peeps, right.

DN: Actually, Peeps didn’t fare well, and we forgot Goldenberg’s Peanut Chews (though someone wrote it in).

BC: And I just realized we didn’t put Hot Tamales on there.

DN: Political results were interesting.

BC: Do tell.

DN: I don’t think we’re telling tales out of school to announce that people prefer Blue M&Ms to Red M&Ms by a 2-to-1 margin. Although to be fair, most folks didn’t seem to care one way or another.

BC: I don’t think we’re telling tales out of school to say that Red folks preferred Skittles more than Blue did.

DN: You're really extrapolating beyond statistical validity, I fear.

BC: I like how now you act like that's a concern.

DN: Speaking more scientifically, people who chose the "Yahoo! Finance" headlines at the bottom preferred cash too. That makes sense.

BC: Plus, BoingBoing readers are overwhelmingly scientifically curious (choice of "Science" in the last question, n=983 out of 1232). That's hopeful.

DN: Yeah, you don't get that scientific anchoring in those off-brand polls, like the Influenster one I saw last week. Besides, I think they invalidate their own survey since Candy Corn was highest rated with their metrics.

BC: Speaking of scientific legitimacy, I can't believe we haven't talked about the results that are already shocking the world.

DN: You're talking about the Friday/Sunday question, I assume.

BC: Of course, Dave, yes, I'm talking about the Friday/Sunday question. Last year we had a near perfect 67:33 ratio of a Friday-to-Sunday preference.

DN: But something happened, because this year we had a sea change, to 65:35 Fri:Sun.

BC: I don't even know what's real anymore. We've continued that survey at Various Breads and Butters for a year now, with lock-tight 2:1 results.

The Platinum Ratio, as people call it now.

DN: It could be bad data. Or campaign fatigue.

BC: Would explain why people are definitely poll-weary, that's something.

DN: Good point. We only had about a fourth the respondents as last year, at close to 1300. It's the political season. People are done with it.

BC: They're done with apple questions too.

DN: Yeah, I'm still struggling to figure out what that East-West apple-eating question is.

BC: Most people are. It just shows that you eat apples from side to side, not bottom to top (core and all).

DN: Why is that even a question?

BC: There's some freak in my hometown that does it that freak way, and then some guy in Northern Virginia. Outliers.

DN: Tell them to eat apples 10,000 times, they'll learn.

BC: Donny P said "I hope they print my manifesto about the correct way to eat an apple," but nimelennar knew the sticking point "I doubt they will. The apple lobby is dominated by in-ciders." This entire line of reasoning has been redacted.

BC: Any other insights to offer.

DN: Not really, but how about lots of graphs.

BC: And more footnotes?

DN: Yes, footnotes galore.

BC: I can't go on like this.

DN: Well? Shall we go?

BC: What?

DN: Shall we go?

BC: Yes, let's go.

They do not move. Cos of the candy? Sugar crash.

There's the stuff about science, and ranking, and surveys, and strata, we say that every year. There's the extra layer about experimental proof, rigorous data analysis, metrics of repeatability. Got it, we got it.

So we're here are again.

Last year was a kind of game changer in the candy ranking industry. We split the survey results between those who were actually planning to trick-or-treat and those basing their answers on memories of childhoods long lost to time. We added this whole other layer of survey questions to gauge the character of survey takers. We promised we would find a legit reason for doing so, but we'd figure it out post facto. Which we did. We contrived a reason post facto. Like, did you know people who like black licorice prefer Sundays over Fridays? No wonder they suck (Sundays and black licorice).

Also, did you know that people who ranked peanut butter and chocolate combo candies higher than mint and chocolate combos are wrong to do so?
Why? Because it's just wrong. Because Mint + Chocolate always goes first. Then Caramel + Chocolate. Then, if you must, Peanut Butter + Chocolate. Stuff like that, that's what we got. No bias or rigging here.

Also, what about that 5000+ people voted with over 500,000 individual preferences? God, imagine if we could harness that energy for something that actually made things better on this crazy blue dot of ours.

Now we return in trying times. We were tempted to focus this survey on Tic Tac and Skittles-heavy asides and sly references. You'd all be like, Oh damn, they just did that. But Halloween is still some time away, and we know by now that campaign references only have a shelf life of about ten days, two weeks tops, so that even today, we bet half of you don't know what we're referencing with deplorable Tic Tacs and Skittles. (Unlike Mint + Chocolate candies, which can keep in the pantry for months and still best Peanut Butter + Chocolate in a straight up two-way race.)

With sophisticated survey tools at our disposal and genuine statistical analysis to follow, here is our third annual Candy Hierarchy survey (that makes this a longitudinal study folks!) and, overall, the preparation for our tenth annual Candy Hierarchy. Please fill it out to the best of your ability, and we'll report back, as always, on Halloween.

The Third Annual Candy Hierarchy Survey

(View this graphic as a huge PDF)

ABSTRACT

It's always about the candy. The Candy Hierarchy is full up with this "joy induction" measurement, this thing that the co-principle investigators (PIs) Cohen and Ng go on about each year. From 2006 to 2013, the PIs conducted a longitudinal study, more or less guided by PI expertise and whim (or whimsical expertise) and possible corporation sponsorship. Research by others in the field sought to refute the findings, obviously unsuccessfully. Yet the PIs were so moved by the yearly outpouring of commentary that they opened up the study to additional data sources, namely people. People who the PIs surveyed. Or is it whom? Anyway, nobody cares – this is about sugar. The 2014 Candy Hierarchy was thus defined by data analysis of 43,767 votes obtained from 1286 individuals. Good for them. But not good enough for science. Because the 2015 Candy Hierarchy doubled down and reworked the whole thing with all kinds of more stuff. This hierarchy therefore presents the newly calculated 2015 rankings, based on a total of 518,605 data points obtained from 5459 individuals in a randomized fashion. It also provides the raw data from a secondary study that sought to understand the character of the survey takers, or rather how character affects joy induction. It's all in there, just go check out the figures.

TRANSCRIPTION OF THIS MORNING'S CONFERENCE PROCEEDING DISCUSSION, WITH DR. COHEN AND DR. NG.

BC: Don't you love how they call us Dr.?

DN: I don't mind.

BC: Well, sure. But you don't struggle with the medical doctor/PhD doctor thing?

DN: Nope. I'm cool with it.

BC:

DN: You're not?

BC: No, no I'm not.

DN:

BC:

DN: Something more you wanted to say?

BC: Listen, I'm squeamish, so even the possibility that someone might ask me for medical help, help that could involve blood, it makes me light headed.

DN: It happens.

BC:

DN: Are we gonna talk candy or do this all morning?

BC: We're gonna talk candy.

BC: Go.

DN: I'm going.

BC:

DN: The hierarchy kind of went through the roof this year.

BC: Way more votes than we knew what to do with, like an order of magnitude more.

DN: Yeah. And the data is open for others to dig into.

BC: Right. More so because there's so much that we didn't have time for.

DN:

BC: Did you see anything surprising?

DN: Meh. I'm not surprised by how the bigger data set (see Figure 1) finally showed dominance of full-sized candy bars over everything else. Proving once again that more is better for trick-or-treaters, gluttons all.

BC: I'm not surprised Dots secured their position in the lower tier.

DN: Right, a lot of chatter on the boards about that too. Dots bring out the kind of anger we used to see with Candy Corn.

BC:

DN:

BC: But you know what was weird?

DN: Do tell.

BC: People prefer Vicodin to Dots.

DN: Not weird at all, Ben, totally unweird. You know what I did think was weird?

BC: The floor is yours.

DN: You still can't accept peanut butter's ascendancy in the past few years.

BC: It's a matter of principle. The public's duped by Big Peanut. Big Mint and
Big Caramel need to lobby more.

DN: What?

BC: Mint > Caramel > Peanut Butter. It goes in that order. All are chocolate combinations, but it's the right combo that matters.

DN: Twix and Snickers were pretty high up. Both caramel-laden.

BC: But Peppermint Patties? Junior Mints? Give them their due. Somehow Nerds were higher than Junior Mints, Dave. Nerds.

DN: So you're disagreeing with 518,605 votes?

BC: It would seem so, Doctor.

DN:

BC:

DN: Let me reset things, this is taking too long. I've got two points: one is the change in the Petersen Inflex (where votes for Joy and Despair cancel each out), which hits right about, ahem, at Vicodin. What does that mean? The other is the ranking pulled from the extra character data.

BC: Did we find anything interesting there?

DN: Maybe? Take a look yourself (Table 1A). You can totally see that the data pulled from kids actually trick or treating provide some interesting insight.

BC: As in the kids really hate kale smoothies. That ranked last.

DN: And also the prominence of fruity chewy things. My God, Skittles is near the top!

BC: It looks like adults hate broken glow sticks the most.

DN: They want ones that work.

BC: That's the only viable conclusion we can reach.

DN: Check out Table 1B too. It kind of looks like joy induction is similar regardless of whether a person chose Betty or Veronica.

BC: Or Friday or Sunday. I'm especially interested in that one. It's a big topic in my circles.

DN: Or even those that listed comic sans as their favourite font, despite prior evidence from this report.

BC:

DN: It's as if Candy Joy Induction is this universal or primal thing.

BC:

DN:

BC: I have no response to that.

DN:

BC:

DN: Can we go now?

BC: I didn't know I was keeping you.

(more…)

(more…)

(more…)

Ranking candy hierarchy is properly and obviously the subject of natural science research. It is based on experimental proof, rigorous data analysis, and the metric of repeatability. This is so obvious that we're not sure why we're even mentioning it. Even so, even so. We're open-minded guys.

So this past year we added in some social science methodology. Gave us something to do on Tuesdays. We had research assistants, undergrads mostly, but also our cousin from Duluth and then his neighbor's son, he's a good kid, he just needs time to get back on his feet, these things happen. The economy. The Mideast. Maybe he shouldn't have been with Kaitlyn from the start. Whatever.

The point is, our research cohort did some oral history work. We did more qualitative research too. We conducted semi-structured interviews, archival investigations, some survey instruments, participant observation, lots of Netflix. More importantly, we did this to vet a new approach this year, where we'll probe the minds of the public and conduct top of the line "multivariate quantitative techniques." That's the type of research goodness that is not only powerful, but on the syllable count, it actually goes to eleven.

So to summarize: As in what's happening then, is that we're culling proper survey data. Then we'll do that goes-to-eleven-syllable-level analysis. It's tremendous this. Seriously, even Kaitlyn would've approved.

(more…)

(more…)