Why do they makes cars (by Jacob Mandel — guest blogger)

bicycle-159796_640

Pollution is bad because it makes people die sometimes, and bikes and scooters don’t make pollution so why do people make cars? It is because if you want to go a long way you have to take a car.Well, bikes can often keep up with cars and scooters can too, so why do cars go faster? It is because they have gas, which makes them go so fast. But it’s the gas that pollutes the world, so we should ask ourselves is it so important to go so fast and so far? No! Because if you want to go so far you can go on a bus or a train, which at least takes a lot people at the same time so it reduces pollution.

Jacob Mandel is five years old and lives in Toronto and he is a very good boy sometimes. He is an aspiring scientist and this is his first publication.

Expedient feedback — certainly not what I expected!

Five months ago I submitted a manuscript for publication at a certain journal I shall not name. Having not heard a peep, last night, I sent off a message asking if everything was okay. I got back a message today that made me laugh. It said:

Thank you for your email. Providing you with expedient feedback regarding your manuscript is important to us. We are currently awaiting for reviewer scores to be returned and will notify you of the Journal‘s decision as soon as all reviewers have submitted their reviews.

While I had hoped for expeditious feedback, I never bargained for expedient feedback. It’s too late for expeditious feedback. A five month wait has already ruled that out no matter how good the feedback might eventually be. As for expediency, I can only imagine what that might entail.

Now for anyone wondering what on earth I’m talking about, some definitions: Screen Shot 2014-10-23 at 10.26.14 PM

Screen Shot 2014-10-23 at 10.26.33 PM

So, I’ve been promised convenient and practical, although improper and immoral, feedback on my paper. Providing such feedback, I’m assured, is important to the journal’s editorial team.

Should I tell the person who wrote to me that they’ve been using the wrong word or should I just leave it alone?

How good are strategic intelligence forecasts?

Screen Shot 2014-09-25 at 10.33.10 PM

This is a freshly pressed post that appeared September 25 (today) in Policy Options magazine. In it, I describe how the 6-year intelligence forecasting accuracy study that Alan Barnes and I undertook developed. A detailed description of that study was published in a recent PNAS article that was subsequently summarized in this piece in The Economist. Note that the Economist referred to a 76% accuracy rate. That’s wrong. It was 94%, as indicated in the PNAS paper and today’s post. The 76% figure was actually the adjusted normalized discrimination index value, which is akin to an adjusted eta-squared value — namely, the proportion of variance in the outcomes explained by the intelligence forecasts.

PESTER More Often! Use Probabilistic Estimation for Sound Theory Evaluation and Revision

Halfway around the world there’s a group of scholars, mainly philosophers, who engage daily in discussions about the future of statistical analysis in the social, cognitive, and, behavioral sciences. There’s Geoff and Fiona and, of course, Neil, the benevolent, who invites metaphorical flies like me to rest on the wall.

For some, like Geoff, I may be more real than metaphorical in my fly-likeness. I’ve pestered him enough lately about ESCI, his handy set of SETs: statistical estimation tools, which help enable the application of what he calls the New Statistics. “Geoff, could you add more lines to the meta-analysis tool — I have 38 studies I’d like to assess, and ESCI stops at 30.” “Geoff, can ESCI handle meta-analyses for correlated designs? … No, that’s okay; I understand, but could you perhaps add that when you get the chance? It would be very, very useful since a lot of what we psychologists do involve correlated designs.” Stuff like that. Flies, after all, aren’t the least bit shy. (The metaphor is imperfect, I realize: flies are known to feast on piles of fecal matter, and I, by no means, wish to imply that my feasting on ESCI means that ESCI is of fecal constitution. I remind you that flies also like many other things that you and I would find quite delicious.)

At any rate, much of the discussion I’ve been overhearing lately has to do with the relationship among concepts such as statistical estimation (the crux of the New Statistics), hypothesis testing, null hypothesis significance testing (or NHST, the pest), and theories in science. Some questions recently posed include, What is hypothesis testing in science without NHST? And, could hypothesis testing proceed effectively with only statistical estimation sans significance testing?

First, let me offer a simple example of hypothesis testing that comes to us straight from the annals of psychology: Peter Wason’s rule discovery task, also known as the 2-4-6 task. The task is a familiar one, so let me just sketch the barest details. The experimenter tells the subject that he has a rule in mind which determines which number triplets are members of a target set and, conversely, which are excluded. He gives the subject an example that fits the rule: {2, 4, 6}. The subject’s task is to discover what rule the experimenter has in mind. To do so, he may generate other triplets and the experimenter will inform him whether the example fits the rule or not. What Wason found was that most subjects generated triplets that fit the hypothesized rule. He mistakenly labelled this behavior confirmation bias, but as Klayman and Ha (1987) later noted, the bias shown is not one of confirming the hypothesis, but one of conforming to the hypothesis, a tendency Klayman and his colleague labelled a positive-test strategy. Positive hypothesis tests, in principle, could yield either confirmatory or disconfirmatory evidence. If I test an instance of what I hypothesize to be the experimenter’s rule and it is shown to be false, then I’ve collected disconfirmatory evidence. Unless, I give it less weight than I would had I learned that it was confirmatory, I cannot legitimately be accused of confirmation bias.

At any rate, I merely wanted to show how hypothesis testing could easily be conducted without NHST. Of course, the task has many convenient attributes that make life easier. The experimenter gives perfectly accurate feedback on each triple tested by the subject, thus ruling out a key source of uncertainty. The rule itself is categorical. There are no fuzzy cases that partially fit the rule, but partially don’t. It’s all black and white. Therefore, if the subject stumbles upon a triplet that doesn’t fit the rule he has in mind, he can be absolutely sure that he has the wrong rule in mind. Accruing confirmatory evidence, however, merely strengthens support for the hypothesis. It never proves the hypothesized rule to be true.

In most behavioral science research, the situation is quite different, mainly because, for a variety of reasons, a single case or even multiple cases of disconfirmatory evidence don’t prove anything. Evidence tends to be tagged with multiple uncertainties. Theories and even hypotheses tend to be fuzzily stated making evidential interpretations more equivocal and subjective. The use of NHST, as the primary means of statistically testing hypotheses, further frustrates the scientific process since it promotes a focus on the null hypothesis, which is not the hypothesis of interest. More damning is the fact that everything but the null hypothesis becomes the alternative! This actively promotes a disregard for precision and clear thinking. It actively promotes satisfaction with merely showing differences between conditions, without thinking carefully about how large or small those differences ought to be based on one’s hypotheses. That, of course, severely undermines scientific progress since the same evidence is often treated by one theoretical camp as supporting a proposition, while an opposing camp treats the same evidence as contradicting the proposition. The wishy-washy-ness of the process actively encourages theoretical dogfights. At least, the camp members see dogfights. From far outside, the fight looks more like two barely living chickens pecking at each other’s broken bodies. To the outsiders, neither chicken appears to be winning and neither looks too wise.

Take an idealized example of the sort of chicken fight behavioral decision theorists love to have. The camps even have fancy names like Meliorists or Pessimists versus Panglossians or Optimists. The former tend to cast human judgment and decision making as irrational, while the former see it as ecologically well-adapted and, well, more or less rational under the circumstances. Let’s say a certain decision-making test is conducted and, by chance, we’d expect a 50% accuracy rate. Of course, quite often, researchers disagree on what accurate means in a given context, but let’s say that all sides agree. Let’s say a sample is collected and found to have a 75% accuracy rate, which given the sample size is significantly different from chance performance (50%). The Optimists say, “see, people are significantly better than chance. There is good evidence for human rationality!” The Pessimists counter, “not so fast! People are significantly worse than the optimal (100%). There is good evidence for human irrationality!”

Now, let’s say we had a confidence interval around the 75% sample estimate, such that we were 95% confident that the accuracy rate was 75% plus or minus 10%. Note that this may do little to stop nearly dead chickens from fighting. Each side can still make essentially the same claims as they would with NHST. However, switching to estimation may do something since with the two-sided confidence interval it may be a bit harder to ignore the other camp’s perspective. Yes, 50% is outside the interval, but so is 100% — and vice versa. So, perhaps an estimative approach would encourage perspective taking a bit more in cases such as these. Maybe, but I won’t hold my breath.

The disagreement in this case stems mainly from the application of differing standards. The Optimists compare human performance to a dart-throwing chimp; the Pessimists compare human performance to Laplace’s demon. They disagree mainly on what the proper benchmark for gauging human performance should be. But, since they are supposed to be empiricists, they pretend to be arguing about the evidence they collected. Secretly, both sides know that performance is better than chance and far from optimal. However, neither side feels it can admit to the full statement, so the unnecessary pecking continues. It’s not even clear if there are imaginary pitbulls in the ring. Perhaps adversarialism is merely a mish-mash of ritualism and strategic calculation.

At any rate, I’m back where I always seem to end up: believing that what science and scientists need most of all is clear, well explicated reasoning that is allowed to flourish. The conditions that promote such flourishing are against ritualism and dogma, gaming the system, and imprecision and deception (including self-deception).

Probabilistic estimation of the kind actively promoted in the New Statistics is a positive step forwards, but it should be for something. For what? Hypothesis testing? That is too narrow an aim. Silly hypotheses may be tested in order to buttress crumbling theories. Sound probabilistic estimation may be recruited in the service of such ignoble aims. Those who are advancing the statistical methods are well placed to remind scientists of that.

No, we need probabilistic estimation to be for more than hypothesis testing. It should, in general, be for theory evaluation and revision. I liked that at the start of this post because it spelled out PETER, and was catchy (get it: Probabilistic Estimation for Theory Evaluation and Revision). But now, I realize it’s sorely incomplete because probabilistic estimation can be used for theory evaluation and revision in ways that are either sound or unsound. Most of what I’ve commented on here is the fact that the soundness of that process is what matters most.

So, while PETER was catchy, all is not lost. I’ve decided to PESTER you by proposing that we use Probabilistic Estimation for Sound Theory Evaluation and Revision.

If using the New Statistics for PESTER can be called pestering, then I wish my Australian colleagues much success pestering scientists and getting them to be pesterers themselves.  I would gladly join them in that cause.

Advice to the Editorial Team at Psychological Science

A few years ago, I sent a manuscript to Psychological Science. I sent it there because it dealt with a problem that I thought would be of interest to a wide audience: the intersection of decision-making, human rationality, linguistics and the intersubjectivity of meaning. The editor who read the manuscript decided to reject it without sending it out for review, even though he said it was “well done” and “interesting”. However, he thought that it was better suited to a specialty journal. Not that I have anything against speciality journals–they are after all the backbone of scholarly publishing. However, when the topic of a paper is especially broad, dealing with methodological, theoretical, and meta-theoretical claims in Nobel prize-winning work, then I have to wonder what decision rule the editor was using.

You see, in the olden days, you sent a paper to a journal and you got a couple of reviews back, along with the editor’s own comments about your paper and perhaps on what to take seriously in the reviewers’ comments. You didn’t always agree with the feedback or like the decision, but at least you had a sense that the paper was peer reviewed. When an editor tells you a paper is well done and interesting, but better off in a specialty journal, you learn nothing other than something about the whims of the editorial staff, at least when you know the paper is quite sweeping.

The journal’s defence of this uninformative process is the same as that of the Old Woman Who Lived in a Shoe. It is simply inundated with so many submissions that it has to reject many without peer review (remember, in the old woman’s case, she simply didn’t know what to do). It only has so many physical paper pages on which to print its selections. Okay, I get that, not that it makes a whole lot of sense in this day and age to publish in print and regulate scientific dissemination on the basis of physical page counts. At least it doesn’t make good scientific sense (and certainly not environmental sense). It might make economic sense, but isn’t APS, which publishes Psychological Science as its flagship journal all about promoting psychology as a science? How do they reconcile that objective with the science distorting policy of a couple of editors gambling on which papers will have the greatest impact for their journal (which will boost their impact factor, thus creating even more pressure for even less transparency in reviewing since they will have to reject even more papers without much consideration). I’m not saying the editors don’t gamble well. The journal’s impact factor is great, so they must be good at picking horses. But, it leaves me wondering how committed to science this approach can be.

At any rate, what really struck me in that rejection letter was the presumption that because they were rejecting my paper that it could only be published in a specialty journal. First the assessment: “We are therefore declining further review of your paper and believe it would be a better fit for a more specialized journal.” Then the quantum of solace: “I am sorry to report such unwelcome news but I hope that the quick decision offers some compensation. Also, I want to mention that, because the journal receives so many manuscripts (nearly 2,600 new submissions last year), roughly two out of three submitted manuscripts are declined during this initial evaluation.”

Okay…. we know you’re popular, but why presume the news is “such unwelcome news”? “Unwelcome news” is presumptuous enough, but such unwelcome news! If the editor had said this unwelcome news, it would have been one thing, but such in this context has a different meaning. I presume editors choose their words carefully, so the presumptuousness was intentional. But maybe I am being too charitable. After all, when I read that the “quick decision” might offer “compensation”, I had to laugh. Really? Is it professional for editors to infer the psychological reactions of authors? I think it is just awful, condescending behaviour. A more appropriate sentence would be “I regret having to reject your manuscript, but I do hope that the rapid turnaround on our review process was at least beneficial in allowing you now to plan your next steps.”  Or something along those lines. In case it’s not clear, “quick decision” might be interpreted as one that was made automatically with little reasoned effort–what is popularly called “System 1” thinking these days, or “thinking fast”–jargon and metaphor, respectively, for automaticity.

As if that wasn’t bad enough, then the rub: “Finally, it’s worth noting that our experience is that manuscripts declined by Psychological Science are typically well received by excellent specialty journals. I wish you much success publishing PSCI-XX-XXXX in such a journal, and I encourage you to continue to consider Psychological Science as an outlet for your work.” Okay, I get it! You really don’t think my paper is worth publishing anywhere else but in a specialty journal. Indeed, you (dear editor) must be so confident in that assessment that you are only willing to wish me luck in publishing in a specialty journal. That is, you have essentially concluded that because you’ve rejected my manuscript in your generalist journal that no generalist journal anywhere would be interested in the paper. Is this the height of overconfidence in your own decision-making? I think such statements are unprofessional of a journal editor, not to mention unnecessary. First, you presume my psychological reactions, then you forecast my ostensibly limited options.   

The paper in question was eventually published in JEP: General. Not exactly a specialty journal, unless you call your speciality psychology.

At any rate, time passes. Then, towards the end of last year, I submitted a manuscript on an entirely different topic to Psychological Science. Amazingly, the editor even sent it out for review. One review was mainly positive; the other mainly negative. The Editor claimed to have “perused” my manuscript as well. Now peruse is an interesting choice of words since it can mean almost opposite things, and it is unclear what meaning this editor intended. It could mean he considered it with attention and in detail, or it could mean he looked it over in a casual or cursory manner. As a small suggestion, I’d recommend less ambiguous terms. Given that the editor did not pick up on a very clear error made by one reviewer (who thought I was analyzing four data bases when in fact I had analyzed five–and, yes, for reasons I won’t go into, that mattered; or at least it would have, had it been true), and indeed repeated the error in his own comments, I must conclude that perused in this instance meant reading in a cursory manner. At any rate, that’s not the main point.

What caught my attention was the penultimate line: “I wish you well with this project and I hope the reviews are useful as you revise the manuscript for a specialty journal.” Whaaaaaat? This was even weirder than last time because the issue of generality-specificity of the topic didn’t even come up in the editor’s or reviewers’ comments. What this also made clear is that the view expressed by the first editor is not an isolated example. Is this Psychological Science’s editorial policy? To assume that the papers they reject would only be acceptable to speciality journals? It seems the editors are not content with rejecting papers from their generalist journal, they implicitly reject authors’ papers from all generalist journals.

Anyway, the paper in question was recently accepted for publication in PLoS ONE. Last time I checked, that journal is a tad more general than Psychological Science.

The reviewing process was also quite a contrast. Instead of telling me how selective they are and where I might be able to publish my manuscript, the editor at PLoS ONE actually said that the paper won’t get published for telling a particularly great story or because of how they forecast its impact. Rather, he said, “What you will need to do is to be very transparent about what your data show and do not show. Please try to be as objective as you can. You will eventually get this paper published in PLOS ONE not for a particularly great story that is hardly supported by the data, but for a scientifically sound study and a similarly sound interpretation of the data.” Wow, isn’t that refreshing? It reminds me of what I’ve always thought science dissemination should be like. The difference in approach raises a much more general issue that deserves a separate post–namely, the economics of science dissemination under the paid subscription hardcopy and open access digital only models, and how those economic models affect the quality of science itself.  

But, here I have a much smaller aim: simply a word of advice to Psychological Science editors about their rejection letters: if you only change one thing, stop assuming that the authors whose papers your reject–namely the roughly 90% you reject–have no other options for publishing to a general audience. At least, stop conveying that assumption openly in the closing statements of your action letters. It is presumptuous, unnecessary, and unprofessional.

Finally–and this is directly to those editors, present and future–if I should give your journal another chance at some point in the future, please do not hold this free advice against me. That too would be unprofessional. Editors, like authors, should be grateful for, and act on, constructive feedback.

 

 

 

The Futility of Significance (Statistical, that is)

A wonderful surprise arrived in my gmail inbox this morning. A message from Neil Thomason, a dear colleague of mine, who sends me interesting things to read from time to time. It’s not that much of the time the things Neil sends me aren’t that interesting. It’s just that the interesting things he does send me only come from time to time.

Last time he sent me something, it was this Many Labs paper on replications available at the Open Science Forum. Clearly, Neil was impressed. Ten of 13 classic effects were well replicated, one was so-so, and two were not replicated. Neil must have thought I was having a bad day when I clearly was not sharing in the joy. The main reason for that has to do with the fact that effects and explanations of effects are seldom properly decoupled. As a result, a replication of a classic effect is often communicated as a replication of the explanation of the effect. But of course an effect can be highly replicable yet explained in an entirely improper manner. I believe that the authors of the Many Labs paper fell into that trap of not adequately separating the effects they replicated from the standard interpretations given to them. I went on in my reply at some length about the inherent dangers of doing so (and I should blog about this in a separate post at some point in the future).

I confess that I was disappointed not to get a reply from Neil since I thought my comments posed a thoughtful reply to his exuberant reaction. I thought those comments would spark a lively discussion of the issue, perhaps leading us toward some sound principles to guide similar future endeavours. Neil is after all a philosopher of science.

At any rate, nearly 3 months later, this morning, I got a message from him. Just this link to a blog post by Matthew Hankins entitled “Still Not Significant.” I really enjoyed the post, which deals with the many creative ways researchers find to describe their results when their statistical significance values are at or greater than the magical p < 0.05 level. I encourage you to read the post, but just in case you’re too busy or lazy to do so, I’ve reproduced the list of circumlocutory expressions quoted there. I warn you, the list is long, but also entertaining:

(barely) not statistically significant (p=0.052)
a barely detectable statistically significant difference (p=0.073)
a borderline significant trend (p=0.09)
a certain trend toward significance (p=0.08)
a clear tendency to significance (p=0.052)
a clear trend (p<0.09)
a clear, strong trend (p=0.09)
a considerable trend toward significance (p=0.069)
a decreasing trend (p=0.09)
a definite trend (p=0.08)
a distinct trend toward significance (p=0.07)
a favorable trend (p=0.09)
a favourable statistical trend (p=0.09)
a little significant (p<0.1)
a margin at the edge of significance (p=0.0608)
a marginal trend (p=0.09)
a marginal trend toward significance (p=0.052)
a marked trend (p=0.07)
a mild trend (p<0.09)
a moderate trend toward significance (p=0.068)
a near-significant trend (p=0.07)
a negative trend (p=0.09)
a nonsignificant trend (p<0.1)
a nonsignificant trend toward significance (p=0.1)
a notable trend (p<0.1)
a numerical increasing trend (p=0.09)
a numerical trend (p=0.09)
a positive trend (p=0.09)
a possible trend (p=0.09)
a possible trend toward significance (p=0.052)
a pronounced trend (p=0.09)
a reliable trend (p=0.058)
a robust trend toward significance (p=0.0503)
a significant trend (p=0.09)
a slight slide towards significance (p<0.20)
a slight tendency toward significance(p<0.08)
a slight trend (p<0.09)
a slight trend toward significance (p=0.098)
a slightly increasing trend (p=0.09)
a small trend (p=0.09)
a statistical trend (p=0.09)
a statistical trend toward significance (p=0.09)
a strong tendency towards statistical significance (p=0.051)
a strong trend (p=0.077)
a strong trend toward significance (p=0.08)
a substantial trend toward significance (p=0.068)
a suggestive trend (p=0.06)
a trend close to significance (p=0.08)
a trend significance level (p=0.08)
a trend that approached significance (p<0.06)
a very slight trend toward significance (p=0.20)
a weak trend (p=0.09)
a weak trend toward significance (p=0.12)
a worrying trend (p=0.07)
all but significant (p=0.055)
almost achieved significance (p=0-065)
almost approached significance (p=0.065)
almost attained significance (p<0.06)
almost became significant (p=0.06)
almost but not quite significant (p=0.06)
almost clinically significant (p<0.10)
almost insignificant (p>0.065)
almost marginally significant (p>0.05)
almost non-significant (p=0.083)
almost reached statistical significance (p=0.06)
almost significant (p=0.06)
almost significant tendency (p=0.06)
almost statistically significant (p=0.06)
an adverse trend (p=0.10)
an apparent trend (p=0.286)
an associative trend (p=0.09)
an elevated trend (p<0.05)
an encouraging trend (p<0.1)
an established trend (p<0.10)
an evident trend (p=0.13)
an expected trend (p=0.08)
an important trend (p=0.066)
an increasing trend (p<0.09)
an interesting trend (p=0.1)
an inverse trend toward significance (p=0.06)
an observed trend (p=0.06)
an obvious trend (p=0.06)
an overall trend (p=0.2)
an unexpected trend (p=0.09)
an unexplained trend (p=0.09)
an unfavorable trend (p<0.10)
appeared to be marginally significant (p<0.10)
approached acceptable levels of statistical significance (p=0.054)
approached but did not quite achieve significance (p>0.05)
approached but fell short of significance (p=0.07)
approached conventional levels of significance (p<0.10)
approached near significance (p=0.06)
approached our criterion of significance (p>0.08)
approached significant (p=0.11)
approached the borderline of significance (p=0.07)
approached the level of significance (p=0.09)
approached trend levels of significance (p0.05)
approached, but did reach, significance (p=0.065)
approaches but fails to achieve a customary level of statistical significance (p=0.154)
approaches statistical significance (p>0.06)
approaching a level of significance (p=0.089)
approaching an acceptable significance level (p=0.056)
approaching borderline significance (p=0.08)
approaching borderline statistical significance (p=0.07)
approaching but not reaching significance (p=0.53)
approaching clinical significance (p=0.07)
approaching close to significance (p<0.1)
approaching conventional significance levels (p=0.06)
approaching conventional statistical significance (p=0.06)
approaching formal significance (p=0.1052)
approaching independent prognostic significance (p=0.08)
approaching marginal levels of significance p<0.107)
approaching marginal significance (p=0.064)
approaching more closely significance (p=0.06)
approaching our preset significance level (p=0.076)
approaching prognostic significance (p=0.052)
approaching significance (p=0.09)
approaching the traditional significance level (p=0.06)
approaching to statistical significance (p=0.075)
approaching, although not reaching, significance (p=0.08)
approaching, but not reaching, significance (p<0.09)
approximately significant (p=0.053)
approximating significance (p=0.09)
arguably significant (p=0.07)
as good as significant (p=0.0502)
at the brink of significance (p=0.06)
at the cusp of significance (p=0.06)
at the edge of significance (p=0.055)
at the limit of significance (p=0.054)
at the limits of significance (p=0.053)
at the margin of significance (p=0.056)
at the margin of statistical significance (p<0.07)
at the verge of significance (p=0.058)
at the very edge of significance (p=0.053)
barely below the level of significance (p=0.06)
barely escaped statistical significance (p=0.07)
barely escapes being statistically significant at the 5% risk level (0.1>p>0.05)
barely failed to attain statistical significance (p=0.067)
barely fails to attain statistical significance at conventional levels (p<0.10
barely insignificant (p=0.075)
barely missed statistical significance (p=0.051)
barely missed the commonly acceptable significance level (p<0.053)
barely outside the range of significance (p=0.06)
barely significant (p=0.07)
below (but verging on) the statistical significant level (p>0.05)
better trends of improvement (p=0.056)
bordered on a statistically significant value (p=0.06)
bordered on being significant (p>0.07)
bordered on being statistically significant (p=0.0502)
bordered on but was not less than the accepted level of significance (p>0.05)
bordered on significant (p=0.09)
borderline conventional significance (p=0.051)
borderline level of statistical significance (p=0.053)
borderline significant (p=0.09)
borderline significant trends (p=0.099)
close to a marginally significant level (p=0.06)
close to being significant (p=0.06)
close to being statistically significant (p=0.055)
close to borderline significance (p=0.072)
close to the boundary of significance (p=0.06)
close to the level of significance (p=0.07)
close to the limit of significance (p=0.17)
close to the margin of significance (p=0.055)
close to the margin of statistical significance (p=0.075)
closely approaches the brink of significance (p=0.07)
closely approaches the statistical significance (p=0.0669)
closely approximating significance (p>0.05)
closely not significant (p=0.06)
closely significant (p=0.058)
close-to-significant (p=0.09)
did not achieve conventional threshold levels of statistical significance (p=0.08)
did not exceed the conventional level of statistical significance (p<0.08)
did not quite achieve acceptable levels of statistical significance (p=0.054)
did not quite achieve significance (p=0.076)
did not quite achieve the conventional levels of significance (p=0.052)
did not quite achieve the threshold for statistical significance (p=0.08)
did not quite attain conventional levels of significance (p=0.07)
did not quite reach a statistically significant level (p=0.108)
did not quite reach conventional levels of statistical significance (p=0.079)
did not quite reach statistical significance (p=0.063)
did not reach the traditional level of significance (p=0.10)
did not reach the usually accepted level of clinical significance (p=0.07)
difference was apparent (p=0.07)
direction heading towards significance (p=0.10)
does not appear to be sufficiently significant (p>0.05)
does not narrowly reach statistical significance (p=0.06)
does not reach the conventional significance level (p=0.098)
effectively significant (p=0.051)
equivocal significance (p=0.06)
essentially significant (p=0.10)
extremely close to significance (p=0.07)
failed to reach significance on this occasion (p=0.09)
failed to reach statistical significance (p=0.06)
fairly close to significance (p=0.065)
fairly significant (p=0.09)
falls just short of standard levels of statistical significance (p=0.06)
fell (just) short of significance (p=0.08)
fell barely short of significance (p=0.08)
fell just short of significance (p=0.07)
fell just short of statistical significance (p=0.12)
fell just short of the traditional definition of statistical significance (p=0.051)
fell marginally short of significance (p=0.07)
fell narrowly short of significance (p=0.0623)
fell only marginally short of significance (p=0.0879)
fell only short of significance (p=0.06)
fell short of significance (p=0.07)
fell slightly short of significance (p>0.0167)
fell somewhat short of significance (p=0.138)
felt short of significance (p=0.07)
flirting with conventional levels of significance (p>0.1)
heading towards significance (p=0.086)
highly significant (p=0.09)
hint of significance (p>0.05)
hovered around significance (p = 0.061)
hovered at nearly a significant level (p=0.058)
hovering closer to statistical significance (p=0.076)
hovers on the brink of significance (p=0.055)
in the edge of significance (p=0.059)
in the verge of significance (p=0.06)
inconclusively significant (p=0.070)
indeterminate significance (p=0.08)
indicative significance (p=0.08)
is just outside the conventional levels of significance
just about significant (p=0.051)
just above the arbitrary level of significance (p=0.07)
just above the margin of significance (p=0.053)
just at the conventional level of significance (p=0.05001)
just barely below the level of significance (p=0.06)
just barely failed to reach significance (p<0.06)
just barely insignificant (p=0.11)
just barely statistically significant (p=0.054)
just beyond significance (p=0.06)
just borderline significant (p=0.058)
just escaped significance (p=0.07)
just failed significance (p=0.057)
just failed to be significant (p=0.072)
just failed to reach statistical significance (p=0.06)
just failing to reach statistical significance (p=0.06)
just fails to reach conventional levels of statistical significance (p=0.07)
just lacked significance (p=0.053)
just marginally significant (p=0.0562)
just missed being statistically significant (p=0.06)
just missing significance (p=0.07)
just on the verge of significance (p=0.06)
just outside accepted levels of significance (p=0.06)
just outside levels of significance (p<0.08)
just outside the bounds of significance (p=0.06)
just outside the conventional levels of significance (p=0.1076)
just outside the level of significance (p=0.0683)
just outside the limits of significance (p=0.06)
just outside the traditional bounds of significance (p=0.06)
just over the limits of statistical significance (p=0.06)
just short of significance (p=0.07)
just shy of significance (p=0.053)
just skirting the boundary of significance (p=0.052)
just tendentially significant (p=0.056)
just tottering on the brink of significance at the 0.05 level
just very slightly missed the significance level (p=0.086)
leaning towards significance (p=0.15)
leaning towards statistical significance (p=0.06)
likely to be significant (p=0.054)
loosely significant (p=0.10)
marginal significance (p=0.07)
marginally and negatively significant (p=0.08)
marginally insignificant (p=0.08)
marginally nonsignificant (p=0.096)
marginally outside the level of significance
marginally significant (p>=0.1)
marginally significant tendency (p=0.08)
marginally statistically significant (p=0.08)
may not be significant (p=0.06)
medium level of significance (p=0.051)
mildly significant (p=0.07)
missed narrowly statistical significance (p=0.054)
moderately significant (p>0.11)
modestly significant (p=0.09)
narrowly avoided significance (p=0.052)
narrowly eluded statistical significance (p=0.0789)
narrowly escaped significance (p=0.08)
narrowly evaded statistical significance (p>0.05)
narrowly failed significance (p=0.054)
narrowly missed achieving significance (p=0.055)
narrowly missed overall significance (p=0.06)
narrowly missed significance (p=0.051)
narrowly missed standard significance levels (p<0.07)
narrowly missed the significance level (p=0.07)
narrowly missing conventional significance (p=0.054)
near limit significance (p=0.073)
near miss of statistical significance (p>0.1)
near nominal significance (p=0.064)
near significance (p=0.07)
near to statistical significance (p=0.056)
near/possible significance(p=0.0661)
near-borderline significance (p=0.10)
near-certain significance (p=0.07)
nearing significance (p<0.051)
nearly acceptable level of significance (p=0.06)
nearly approaches statistical significance (p=0.079)
nearly borderline significance (p=0.052)
nearly negatively significant (p<0.1)
nearly positively significant (p=0.063)
nearly reached a significant level (p=0.07)
nearly reaching the level of significance (p<0.06)
nearly significant (p=0.06)
nearly significant tendency (p=0.06)
nearly, but not quite significant (p>0.06)
near-marginal significance (p=0.18)
near-significant (p=0.09)
near-to-significance (p=0.093)
near-trend significance (p=0.11)
nominally significant (p=0.08)
non-insignificant result (p=0.500)
non-significant in the statistical sense (p>0.05
not absolutely significant but very probably so (p>0.05)
not as significant (p=0.06)
not clearly significant (p=0.08)
not completely significant (p=0.07)
not completely statistically significant (p=0.0811)
not conventionally significant (p=0.089), but..
not currently significant (p=0.06)
not decisively significant (p=0.106)
not entirely significant (p=0.10)
not especially significant (p>0.05)
not exactly significant (p=0.052)
not extremely significant (p<0.06)
not formally significant (p=0.06)
not fully significant (p=0.085)
not globally significant (p=0.11)
not highly significant (p=0.089)
not insignificant (p=0.056)
not markedly significant (p=0.06)
not moderately significant (P>0.20)
not non-significant (p>0.1)
not numerically significant (p>0.05)
not obviously significant (p>0.3)
not overly significant (p>0.08)
not quite borderline significance (p>=0.089)
not quite reach the level of significance (p=0.07)
not quite significant (p=0.118)
not quite within the conventional bounds of statistical significance (p=0.12)
not reliably significant (p=0.091)
not remarkably significant (p=0.236)
not significant by common standards (p=0.099)
not significant by conventional standards (p=0.10)
not significant by traditional standards (p<0.1)
not significant in the formal statistical sense (p=0.08)
not significant in the narrow sense of the word (p=0.29)
not significant in the normally accepted statistical sense (p=0.064)
not significantly significant but..clinically meaningful (p=0.072)
not statistically quite significant (p<0.06)
not strictly significant (p=0.06)
not strictly speaking significant (p=0.057)
not technically significant (p=0.06)
not that significant (p=0.08)
not to an extent that was fully statistically significant (p=0.06)
not too distant from statistical significance at the 10% level
not too far from significant at the 10% level
not totally significant (p=0.09)
not unequivocally significant (p=0.055)
not very definitely significant (p=0.08)
not very definitely significant from the statistical point of view (p=0.08)
not very far from significance (p<0.092)
not very significant (p=0.1)
not very statistically significant (p=0.10)
not wholly significant (p>0.1)
not yet significant (p=0.09)
not strongly significant (p=0.08)
noticeably significant (p=0.055)
on the border of significance (p=0.063)
on the borderline of significance (p=0.0699)
on the borderlines of significance (p=0.08)
on the boundaries of significance (p=0.056)
on the boundary of significance (p=0.055)
on the brink of significance (p=0.052)
on the cusp of conventional statistical significance (p=0.054)
on the cusp of significance (p=0.058)
on the edge of significance (p>0.08)
on the limit to significant (p=0.06)
on the margin of significance (p=0.051)
on the threshold of significance (p=0.059)
on the verge of significance (p=0.053)
on the very borderline of significance (0.05<p<0.06)
on the very fringes of significance (p=0.099)
on the very limits of significance (0.1>p>0.05)
only a little short of significance (p>0.05)
only just failed to meet statistical significance (p=0.051)
only just insignificant (p>0.10)
only just missed significance at the 5% level
only marginally fails to be significant at the 95% level (p=0.06)
only marginally nearly insignificant (p=0.059)
only marginally significant (p=0.9)
only slightly less than significant (p=0.08)
only slightly missed the conventional threshold of significance (p=0.062)
only slightly missed the level of significance (p=0.058)
only slightly missed the significance level (p=0·0556)
only slightly non-significant (p=0.0738)
only slightly significant (p=0.08)
partial significance (p>0.09)
partially significant (p=0.08)
partly significant (p=0.08)
perceivable statistical significance (p=0.0501)
possible significance (p<0.098)
possibly marginally significant (p=0.116)
possibly significant (0.05<p>0.10)
possibly statistically significant (p=0.10)
potentially significant (p>0.1)
practically significant (p=0.06)
probably not experimentally significant (p=0.2)
probably not significant (p>0.25)
probably not statistically significant (p=0.14)
probably significant (p=0.06)
provisionally significant (p=0.073)
quasi-significant (p=0.09)
questionably significant (p=0.13)
quite close to significance at the 10% level (p=0.104)
quite significant (p=0.07)
rather marginal significance (p>0.10)
reached borderline significance (p=0.0509)
reached near significance (p=0.07)
reasonably significant (p=0.07)
remarkably close to significance (p=0.05009)
resides on the edge of significance (p=0.10)
roughly significant (p>0.1)
scarcely significant (0.05<p>0.1)
significant at the .07 level
significant tendency (p=0.09)
significant to some degree (0<p>1)
significant, or close to significant effects (p=0.08, p=0.05)
significantly better overall (p=0.051)
significantly significant (p=0.065)
similar but not nonsignificant trends (p>0.05)
slight evidence of significance (0.1>p>0.05)
slight non-significance (p=0.06)
slight significance (p=0.128)
slight tendency toward significance (p=0.086)
slightly above the level of significance (p=0.06)
slightly below the level of significance (p=0.068)
slightly exceeded significance level (p=0.06)
slightly failed to reach statistical significance (p=0.061)
slightly insignificant (p=0.07)
slightly less than needed for significance (p=0.08)
slightly marginally significant (p=0.06)
slightly missed being of statistical significance (p=0.08)
slightly missed statistical significance (p=0.059)
slightly missed the conventional level of significance (p=0.061)
slightly missed the level of statistical significance (p<0.10)
slightly missed the margin of significance (p=0.051)
slightly not significant (p=0.06)
slightly outside conventional statistical significance (p=0.051)
slightly outside the margins of significance (p=0.08)
slightly outside the range of significance (p=0.09)
slightly outside the significance level (p=0.077)
slightly outside the statistical significance level (p=0.053)
slightly significant (p=0.09)
somewhat marginally significant (p>0.055)
somewhat short of significance (p=0.07)
somewhat significant (p=0.23)
somewhat statistically significant (p=0.092)
strong trend toward significance (p=0.08)
sufficiently close to significance (p=0.07)
suggestive but not quite significant (p=0.061)
suggestive of a significant trend (p=0.08)
suggestive of statistical significance (p=0.06)
suggestively significant (p=0.064)
tailed to insignificance (p=0.1)
tantalisingly close to significance (p=0.104)
technically not significant (p=0.06)
teetering on the brink of significance (p=0.06)
tend to significant (p>0.1)
tended to approach significance (p=0.09)
tended to be significant (p=0.06)
tended toward significance (p=0.13)
tendency toward significance (p approaching 0.1)
tendency toward statistical significance (p=0.07)
tends to approach significance (p=0.12)
tentatively significant (p=0.107)
too far from significance (p=0.12)
trend bordering on statistical significance (p=0.066)
trend in a significant direction (p=0.09)
trend in the direction of significance (p=0.089)
trend significance level (p=0.06)
trend toward (p>0.07)
trending towards significance (p>0.15)
trending towards significant (p=0.099)
uncertain significance (p>0.07)
vaguely significant (p>0.2)
verged on being significant (p=0.11)
verging on significance (p=0.056)
verging on the statistically significant (p<0.1)
verging-on-significant (p=0.06)
very close to approaching significance (p=0.060)
very close to significant (p=0.11)
very close to the conventional level of significance (p=0.055)
very close to the cut-off for significance (p=0.07)
very close to the established statistical significance level of p=0.05 (p=0.065)
very close to the threshold of significance (p=0.07)
very closely approaches the conventional significance level (p=0.055)
very closely brushed the limit of statistical significance (p=0.051)
very narrowly missed significance (p<0.06)
very nearly significant (p=0.0656)
very slightly non-significant (p=0.10)
very slightly significant (p<0.1)
virtually significant (p=0.059)
weak significance (p>0.10)
weakened..significance (p=0.06)
weakly non-significant (p=0.07)
weakly significant (p=0.11)
weakly statistically significant (p=0.0557)
well-nigh significant (p=0.11)

One blogger even went so far as to implement the list in R so that when a significance level between .05-.12 is registered, it randomly selects one of those “p excuses” to describe the result.

Hankins’ advice is not to waffle: if your p value is below 0.05 (or presumably whatever you set your alpha level to), your result is significant. If it’s at or over that value, it’s not significant. No “marginal” in betweens or creative expressions such as “very closely brushed the limit of statistical significance (p=0.051).” By the way, does anyone not see a counterfactual thinking study here just waiting to be exploited? (Recall “The loser that almost won” by Kahneman and Varey 1990).

In a related blog post, Hankins shows this wonderful figure he generated from Google Scholar search results.

It certainly makes a point. We have a problem when p = 0.10 is described as marginally significant more frequently than values between 0.050 and 0.094.

And Hankins’ comments in the figure essentially capture the psychology of significance test reporting.

But something about the advice just doesn’t make sense to me. Hankins, like most of us, realizes that the 0.05 level is arbitrary. How could it be, then, that it makes sense to describe one’s results as significant when p is just below that arbitrary value — say 0.049 — while an ever so slightly different result — say 0.050 — ought to be instead labelled not significant.

Is this not the path to madness?

Perhaps the use of “marginally significant” is also partly an attempt by researchers to cope with the craziness of an arbitrary cutoff that defines whether they are to claim their results are “significant” or not.

Marginality at least allows a grey zone around the point of insanity to be defined. The problem is not the grey zone, but rather the use of the arbitrary cutoff.

As I noted in my reply to Neil, although more journal editors and reviewers are calling for effect sizes, most still expect some significance testing to accompany that.

Perhaps it would be wiser to stop calling the p values we standardly report “significance” values, and instead call them what they are: the probability of data given the null hypothesis is true.

Doing that would have two immediately beneficial effects. First, it would correctly define those values, which are often misinterpreted as 1 – the probability of the experimenter’s hypothesis being true, given the data. For instance, one commenter states “All a p-value of 0.05 means is there’s a 95% chance that the hypothesis was indeed working.” Second, it would get us away from the arbitrariness of p < 0.05.

In the end, the circumlocution researchers use in describing the results of their significance tests may not be excusable, but it is understandable not only as a tactical maneuver to keep studies out of the unpublished file drawer, but also as a means of coping with the senseless arbitrariness of a point that separates a continuum into two alternatives: significant or not.

Do Quantitative Forecasters Have Special Obligations to Policy Advisees?

A recent Dart-Throwing Chimp blog post by Jay Ulfelder asks the question, how circumscribed should quantitative forecasters be? The question was prompted by recent comments he made at a meeting on genocide where he described his efforts to help build a quantitative system for early warning of genocide. As he notes, “The chief outputs of that system are probabilistic forecasts, some from statistical models and others from a “wisdom of (expert) crowds” system called an opinion pool.” His post was prompted by a set of online replies from one of the other panelists, Patrick Ball, executive director of Human Rights Data Analysis Group.

The gist of Ball’s replies (as summarized by Ulfelder) is that forecasters should be wary of using quantitive techniques in place of more conventional qualitative approaches because policy makers (or other decision makers reliant on forecasting advice) are disproportionately swayed by quantitative information, perhaps especially when it’s visualized as a figure or graph. Such information can, in Ball’s view, crowd out the more conventional forms of human assessment, which Ball sees as having much value. As well, since many users don’t have the technical skills to judge the integrity of the quantitative techniques employed on their own, there is a special obligation, in Ball’s view, for presenting the limitations of quantitative approaches up front to users.

Ulfelder is not convinced. He cites Kahneman (and, by the way, who doesn’t?), who notes that people have a strong bias for human judgment and advice over machine or technology. We take greater pride in human triumphs than in the triumphs of machines (that we built!) and we are also more willing to accept human error than machine or quantitative modelling error. This is why we resist greater reliance on quantitative modelling techniques as judgment aids even when the evidence clearly indicates that judgment accuracy is improved. We just don’t trust machines to inform us like we do humans, even when they do much, much better.

Ulfelder draws on Kahneman in raising another point, which is that advisers often get ahead by inflating their confidence, often way past the point of proper calibration. Hemming and hawing to policy makers about the limitations of one’s quantitative approach is a surefire recipe for advice neglect, especially if it’s done up front as Ball suggests. By the time the advisor gets to the message, the audience may have tuned out, not only because the technical details aren’t what they wanted to know about, but also because the messenger has decided the first thing to tell them about is the problems. That negatively primes the receiver. Since qualitatively-oriented advisors don’t bend over backwards to qualify the limitations of their approaches — the predominant one being “expert intuition” — why should the quantitative advisor self-handicap?

Here’s my take: First, Ball has got a point. We should be concerned that our models or other quantitative approaches to advice giving (forecasts or otherwise) are sound. But, who says developers of such models like Ulfelder aren’t? It seems odd to presume that quantitative types would be less concerned about rigour than their qualitative counterparts. I would have thought that model developers would be more sensitized to issues about cross validation than qualitative types would be to validation or reliability tests, if for no other reason than in the former case there are pretty clear methods for validating and testing reliability, whereas as one moves toward “expert intuition” the methods are murkier. That murk I would think would translate into “rigour neglect.”

Second, Ball has got another point. Many users won’t understand the mechanics behind the model. If they are technophiles, they may unduly trust (which is what I believe he was emphasizing). If they’re technophobes, they may unduly dismiss (which tracks with Ulfelder’s experiences). Either way, their reactions are driven more by their attitudes toward technology and quantification than by accuracy, diagnostic value, relevance, timeliness and other criteria that matter for effective decision making. There is a real problem here because in many cases it’s very hard to explain how the model works, so purveyors may end up saying “just trust me — it works, at least better than the alternatives.” Now advisees will likely fall back on their attitudes. Technophiles might be more inclined to trust, technophobes to doubt.

I don’t think Ball’s solution of pre-emptive warning is the right one, but I do think he was onto something, and that is that it would be very helpful if quantitative forecasters could find a way in layman’s terms to explain their basic approach — what their model does, how it does it, and how we know if it’s any good. That I believe would help foster trust. It’s not a special obligation in my view, but rather a benefit to all sides (except perhaps anyone who feels threatened by the prospect of models replacing humans as sources of advice).

From the quantitative forecaster’s perspective, I’d say this is a strategic necessity because, as Sherman Kent noted long ago (e.g., in words of estimative probability), not only are most assessors “poets” rather than “mathematicians”, but most policy makers are also poets, perhaps even in greater proportion than within the intelligence community. Kent was of course referring to the qualitative types who put narrative beauty ahead of predictive accuracy (poets) and their counterparts in the intelligence community who not only want predictive accuracy but also want to communicate judgements very clearly, preferably with numbers rather than words (mathematicians). To put it into the psychological terms Phil Tetlock articulated some decades ago, the quantitative forecaster is accountable to a skeptical audience and, accordingly, he may engage in some pre-emptive self criticism to show his audience that he has considered all sides. I believe this is, more generally, why strategic analysts are underconfident rather than overconfident in their forecasts, despite showing very good discrimination skill, as we show in a recent report and in a recent paper in the Proceedings of the National Academy of Sciences.

The poets have the home court advantage because they are advising other poets, who are skeptical about the products mathematicians offer. Because of this ecology of beliefs, confidence peddling, which might work quite well for poets, will probably flop for mathematicians, maybe even as badly as up-front self-handicapping. Just as the mathematicians strive for clarity and crispness in forecast communication, they need to do likewise regarding communications about their methods. That’s seldom the case. Too often, the entry price for understanding is set far too high. That puts off even those who might have been inclined to listen to advice from new, more quantitatively-oriented advisors.

And, of course we should be asking about the qualitative human assessments — how good are they? how can we know? — just as Tetlock had done in his landmark study of geopolitical forecasting and as we’ve more recently done with strategic intelligence analysts in the aforementioned report and paper. When forecasters refuse to give forecasts that are verifiable either because their uncertainties are shrouded in the vagueness of verbal probability terms or because the targets of their forecasts are ill defined, it becomes difficult, if not impossible, to verify accuracy. Some poets might like it that way. Their bosses and their bosses’ political masters aren’t going to force them to do it differently since they are mainly poets as well. The mathematicians have to try to advance the accountability issue. It might help if some of them got into high-ranking policy positions. Then again, maybe they don’t make good leaders. The ecology of these individual differences — poets vs. mathematicians, foxes vs. hedgehogs, etc. — across functional roles in society is surely not accidental.