Posts tagged “statistics”

ChittahChattah Quickies

  • [from julienorvaisas] Problem With Studies: Why They Get It Wrong [ABC News] – [Food for thought as we analyze, synthesize, prioritize and report our own study results.] People, including journal editors, naturally prefer papers announcing or at least suggesting a dramatic breakthrough to those saying, in effect, "Ehh, nothing much here." The availability error, the tendency to be unduly influenced by results that, for one reason or another, are more psychologically available to us, is another factor. Results that are especially striking or counterintuitive or consistent with experimenters' pet theories also more likely will result in publication. Even such a prosaic occurrence as clock-watching provides illustration of this. I don't think I look at the clock more than others do, but I always seem to notice and remember when the time is 12:34, but not when it's 10:56 or 7:41.
  • [from steve_portigal] Making culture, provoking culture (by making pie) [Grant McCracken] – [Grant sees in design-provocations the possibility to go beyond wishful-social-change-thinking and uncover rich new insights about people and culture. Plus, pie. Mmm, pie.] Our interest is sincere. We do want to know about them. We want to seize this opportunity to find out who they are, to listen to anything they're prepared to tell us. We are doing ethnography in tiny bite size bits. (Here too we want to consult the work on methodology.) Some people who wish to make a social difference don't really care to hear from the Pie recipient. They have a vision of the new world, and they mean to keep banging away at this vision until the pie recipient embraces it. But if we have learned anything about engaging the world it is that it can't be about us. Our best efforts must begin with a study of them.

ChittahChattah Quickies

  • Shoddy Pear Analytics study of Twitter content – This is really bad research, but it's topical so it'll get press.

    Bad research #1: When creating content categories for your analysis, don't be so dismissive. They established a category for this analysis called "pointless babble" so obviously they have a strong point of view of the value of that content and it's hardly unbiased research.

    Bad research #2: Understand that Twitter (or any social, conversational medium) is a system. It's not an independent set of messages, it's a multifaceted conversation. Where's the analysis to look at the interrelationships and dependencies between "I am eating a sandwich" and "here's a great news story to check out"? That's much more difficult, mathematically, and to even consider that would require deeper insight than Pear showed here.

ChittahChattah Quickies

  • New Yorker profile of Fred Franzia, the rather unpleasant character behind Charles Shaw wine (akaTwo Buck Chuck) – "You tell me why someone's bottle is worth eight dollars and mine's worth two dollars," he says. "Do you get forty times the pleasure from it?" With Two Buck Chuck, Franzia invented a category known as “super-value” wine. Cheap wine – so-called skid-row wine – is noting new; Franzia's idea was to make cheap wine that yuppies would feel comfortable drinking. He put Charles Shaw in a seven-hundred-and-fifty millilitre glass bottle, with a real cork, and used varietal grapes.
  • Offline, accurate quantitative usage data can be tough to capture – Advertisers rely on M.R.I.’s research. It measures how many readers a magazine has, including people who did not buy it but read a friend’s copy or flipped through it at the doctor’s office. It also profiles the readers of all the magazines, including their income levels, attitudes and toothpaste-buying habits.

    M.R.I. divides the country into representative neighborhoods and sends researchers to the zones to conduct a 45-minute interview, with 26,000 people a year, asking them to remember which magazines they have read in the last six months.

    The researchers leave behind a 104-page survey about what sort of television shows people watch, what kind of products they use, and what social or behavioral traits describe them. M.R.I. then tries to adjust its results so they represent the country.

    [But there are accuracy issues] While M.R.I. said Tennis magazine’s readership dropped almost by a third, its subscriptions and newsstand sales rose slightly.

ChittahChattah Quickies

  • PETA (hopefully tongue-in-cheek) attempts to rebrand fish as "Sea Kittens" – Sorta reductio ad absurdum re: my latest interactions column, Poets, Priests, and Politicians
  • Rug company Nanimarquina brings global warming to your living room – "If there is an iconic image that represents the natural devastation of global warming, it is the lone polar bear stuck on a melting ice flow. Now eco rug company Nanimarquina has teamed up with NEL artists to create a beautiful ‘Global Warming Rug’ – complete with stranded polar bear floating in the middle of the sea – to represent the most pressing issue of our time. Rugs have been traditionally used throughout the ages to tell stories and communicate messages, and we think this is a lovely, poignant new take on a time-honored tradition." What effect does it have when an issue like global warming gets iconified and aestheticized like this? Does it drive home the seriousness of the situation, or make it more palatable?
  • Asch conformity experiments – (via Eliezer Yudkowsky) Asch asked people about similarity of height between several lines. Confederates answered incorrectly and this influenced the subject themselves to support this incorrect answer.
  • Confirmation bias: the tendency to seek out information that supports what we already believe – (via Eliezer Yudkowsky) The 2-4-6 problem presented subjects with 3 numbers. Subjects were told that the triple conforms to a particular rule. They were asked to discover the rule by generating their own triples, where the experimenter would indicate whether or not the triple conformed to the rule. While the actual rule was simply “any ascending sequence”, the subjects often proposed rules that were far more complex. Subjects seemed to test only “positive” examples—triples the subjects believed would conform to their rule and confirm their hypothesis. What they did not do was attempt to challenge or falsify their hypotheses by testing triples that they believed would not conform to their rule.
  • Overcoming Bias – Blog by Eliezer Yudkowsky and others about (overcoming) biases in perception, decisions, etc.
  • Hindsight bias: when people who know the answer vastly overestimate its predictability or obviousness, – (via Eliezer Yudkowsky)
    Sometimes called the I-knew-it-all-along effect.
    "…A third experimental group was told the outcome and also explicitly instructed to avoid hindsight bias, which made no difference."
  • Planning fallacy – the tendency to underestimate task-completion times – (via Eliezer Yudkowsky) Asking people what they did last time turns out to be more accurate than what they either hope for or expect to happen this time
  • Cognitive Biases in the Assessment of Risk – (via Eliezer Yudkowsky) Another example of extensional neglect is scope insensitivity, which you will find in the Global Catastrophic Risks book. Another version of the same thing is where people would only pay slightly more to save all the wetlands in Oregon than to save one protected wetland in Oregon, or people would pay the same amount to save two thousand, twenty thousand, or two hundred thousand oil-stroked birds from perishing in ponds. What is going on there is when you say, “How much would you donate to save 20,000 birds from perishing in oil ponds,” they will visualize one bird trapped, struggling to get free. That creates some level of emotional arousal, then the actual quantity gets thrown right out the window.

    [I am not sure that's the reason why; I think there could be other explanations for the flawed mental model that leads to those responses]

  • Conjunction fallacy – (via Eliezer Yudkowsky) A logical fallacy that occurs when it is assumed that specific conditions are more probable than a single general one. Example: Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.

    Which is more probable?

    1. Linda is a bank teller.
    2. Linda is a bank teller and is active in the feminist movement.

    85% of those asked chose option 2 [2]. However, mathematically, the probability of two events occurring together (in "conjunction") will always be less than or equal to the probability of either one occurring alone.


About Steve