Posts

, ,

Musings about statistics by a statistics-phobe

By B. Staffan Lindgren, Professor Emeritus

A while back, a paper accepted by The American Statistician entitled “The ASA’s statement on p-values: context, process, and purpose” was posted to the American Statistical Association website. The gist of the paper was that many disciplines rely too much on the p-value as the sole indicator of research importance. Not surprisingly, the paper received considerable attention.

Over my career, I had a love-hate relationship with statistics, knowing just enough to be dangerous, but not enough to really understand what I was doing. Consequently I relied on packaged software and/or colleagues or students who were more quantitatively minded than myself. For example, I generally made sure that a graduate student committee had at least one member with some strength in statistics to make sure I would not leave the candidate stranded or led astray. So if you read my thoughts below, keep in mind that I tread on very thin ice here. I fully expect some disagreement on this, but that is the way it is supposed to be. Ultimately it is your responsibility to understand what you are doing.

The approaches and tools for statistical analysis have changed a lot since my student days, which was at the dawn of mainframe computers for general use, on which we could use a software package called Textform rather than typing the thesis on a type writer as I (read “a secretary I hired and almost drove to depression”) did for my masters. My first visit to a statistical consultant at Simon Fraser University ended with the advice that “This data set can’t be analyzed, it contains zero values.” The software of choice was SPSS, which did not allow for any complexity, so I did a fair bit by hand (which might have been a good thing since it forced me to think about what I was doing, but certainly did not prevent errors). Later in my career it was sometimes a struggle to decide among differing opinions of statisticians what was and was not appropriate to use, but with a little help from my friends I think I managed to negotiate most of the pitfalls (no pun intended) fairly well.

The author with his eponymous insect trap, sometime after struggles doing statistics with room-sized computers. Photo: Ron Long

The statistic-phobic author with his eponymous insect trap, preparing to gather data and test hypotheses. Photo: Ron Long.

One of the issues with our reliance on p-values is that it is tempting to do post-hoc “significance-hunting” by using a variety of approaches, rather than deciding a priori how to analyze the data. Data that show no significance often remains unpublished, leading to potential “publication bias”. In part this may be the result of journal policies (or reviewer bias), which tends to lead to rejection of papers reporting ‘negative’ results. We have also been trained to use an experiment-wise alpha of 0.05 or less, i.e., a significant result would be declared if the p-value is lower than 0.05. There are two problems with this. First, it is an arbitrary value in a sense, e.g., there really is no meaningful difference between p=0.049 and 0.051. Furthermore, the p-value does not really tell you anything about the importance of the result. All it can do is give some guidance regarding the interpretation of the results relative to the hypothesis. I have tried to make students put their research in context, because I believe the objective of the research may dictate whether or not a significant p-value is important or not. I used to work in industry, and one of the reasons I left was that recommendations I made based on research were not always acted upon. For example, pheromones of bark beetles are often synergized by various host volatiles. But whether or not they are may depend on environmental factors. For example, just after clear cutting the air is likely to have high levels of host volatiles, thus making any host volatile added to a trap ineffective. However, a company may make money by selling such volatiles, and hence they would tend to ignore any results that would lead to a loss of revenue. On the other hand, one could argue that they have the customers’ best interest in mind, because if host volatiles are important under some circumstances, it would be detrimental to remove them from the product.

This leads to my thoughts about the power of an analysis. The way I think of power is that it is a measure of the likelihood of finding a difference if it is there. There are two ways of increasing power that I can think of. One is to increase the number of replications, and the other is to use a higher alpha value. It is important to think about the consequence of an error. A Type I error is when significance is declared when there is none, while a Type II error is when no significance is found when in fact there is one. Which of these is most important is something we need to think about. For example, if you worked in conservation of a threatened species, and you found that a particular action to enhance survival resulted in a p-value of 0.07, would you be prepared to declare that action ineffective assuming that it wasn’t prohibitively expensive? If you have committed a Type II error, and discontinue the action, it could result in extinction of the threatened species. On the other hand, if you test a pesticide, would a significant value of 0.049 be enough to decide to pursue the expensive testing required for registration? If you have committed a Type I error, the product is not likely to succeed in the market place. If the potential market is small, which tends to be the case for behavioural chemicals, it may not be feasible to use this product because of the high cost, which has nothing to do with statistical analysis, but could be the overriding concern in determining the importance of the finding.

One area where the sole use of p-values can become very problematic is for regressions. The p-value only tells us whether or not the slope of the line is significantly different from zero, and therefore it becomes really important to look at how the data are distributed. An outlier can have a huge impact, for example (see figure). As an editor I saw many questionable regressions, e.g., with single points driving much of the effect, but which in the text were described as highly significant.

Fig. 1. An example of where a single point is driving a linear regression. Take it away and there is no apparent relationship at all. Figure from http://www.stat.yale.edu/Courses/1997-98/101/linreg.htm

Finally, we need to keep in mind that a significant p-value does not indicate certainty, but probability, i.e., at p=0.05, you would expect to get the same result 19 of 20 times, but that still means that the result could be the result of chance if you only ran the experiment once. (If you run a biological experiment that yields a p-value close to 0.05 a number of times, you would soon discover that it can be difficult to get the same outcome every time). Depending on the context, that may not be all that confidence inspiring. For example, if someone told you that there was only a 5% probability that you would be get seriously sick by eating a particular mushroom, wouldn’t that make you think twice about eating it?? On the other hand many of us will gladly shell out money to buy a 6/49 ticket even though the probability of winning anything at all is very low, let alone winning the jackpot, because in the end we are buying the dream of winning, and a loss is not that taxing (unless you gamble excessively of course). I consider odds of 1:8000 in a lottery really good, which they aren’t of course, evidenced by the fact that I have never won anything of substance! So relatively speaking, 1:20 is astronomically high if you think about it!

Why am I bothering to write this as a self-confessed statistics phobe? I have mainly to emphasize that you (and by “you” I primarily mean students engaged in independent research) need to think of statistics as a valuable tool, but not as the only, or even primary tool for interpreting results. Ultimately, it is the biological information that is important.

, , , ,

ESC Public Encouragement Grants

By Staffan Lindgren

———–

Given a stimulating, or even neutral environment, I firmly believe that the natural interest that most, if not all kids have for animals will be retained for life – once a bug nerd always a bug nerd. My parents’ photos of me as a kid almost always show me on all four turning rocks or peering down into some pond for any sign of life, and I’m not much different now (See photographs). I was fortunate to grow up close to nature with parents who pretty much put up with anything that I dragged home, usually alive (my mom drew the line at snakes, but spiders were OK as long as they were contained). Many kids are not as fortunate. They may grow up in a large city, or have one or two parents who at best believe that the only good bug is a dead bug and at worst go catatonic at the sight of even the tiniest spider.

The author (left) and a friend looking for moon jellies on the Baltic Sea coast, Arkösund, Sweden, 1958 (Photo R. Lindgren).

The author (left) and a friend looking for moon jellies on the Baltic Sea coast, Arkösund, Sweden, 1958 (Photo R. Lindgren).

Part of the mandate of the ESC and its affiliate regional societies is to stimulate an interest in entomology through education. One of the tools by which ESC tries to do this is by making available a small grant available annually to the regional societies. The ESC committee guidelines on these “Public Encouragement Grants” state:

(a) Each Affiliate shall be eligible to apply to the Committee for an annual grant of $200 for public education.

(b) Application for the Grant shall be made annually to the Chair and should indicate how the money is to be used.

(c) Funds may be accumulated for a maximum period of three years (i.e., up to $600).

(d) Applications in excess of $200 shall be considered.

(d) The difference between the amount made available annually by the Society for public encouragement, and the amount given to Affiliates, shall be called the Public Encouragement Discretionary Fund. This Fund may be used to augment grants to Affiliates or finance direct public encouragement activities of the Committee.

In 2012, two such grants were awarded, one to Société d’entomologie du Québec (SEQ), who reported:

The $200 granted to the SEQ by the ESC for public education and outreach were given to the AEAQ (Association des entomologistes amateurs du Quebec). …They used the funds to grant free access to their annual meeting (6-7 July 2012) to participants of less than 18 yrs old that could not afford registrations. The annual meeting of the AEAQ is primarily a field meeting with some presentations, and a number of interesting faunistic records were made this year. A recap of this meeting with pictures can be found in the bulletin of the AEAQ at the following url: http://aeaq.ca/nouvl/nouvailes222automne2012.pdf

The second grant was made to the Entomological Society of Manitoba (ESM), who reported:

As part of the Youth Encouragement and Public Education Committee activities, members of the ESM travel to visit schools and various special interest groups to talk about insects, and there are many groups that visit the facilities of the Department of Entomology as well to see the live insects and to learn about insects. The Program received $200 from the ESC this year, and this money was used in part for some minor renovations to the insectary (see photograph) in the Department of Entomology. There is now space for new colonies, and the rearing room can now accommodate small groups for tours. The grant was also used to print business cards, which are provided to the different youth groups visited, or to interested parties during festivals where the Youth Encouragement Committee is present.

University of Manitoba insectary after upgrade partially funded by ESC Public Encouragement Grant (Photo: Matt Yunik)

University of Manitoba insectary after upgrade partially funded by ESC Public Encouragement Grant (Photo: Matt Yunik)

It may not appear that activities like these have much impact, but sometimes it takes very little to stimulate the minds of young people. Providing access for youth to meetings allows them to interact with entomologists and learn that we are people too! I was profoundly influenced by the kind, patient, and carefully typed replies (on official Uppsala University letterhead) when I as a 13-year-old confined live spiders in matchboxes and sent these in regular letters by snail-mail to the arachnologist Dr. Åke Holm! (He did tell me that the spiders generally did not arrive in very good condition, incidentally). I have never forgotten that, and I try to treat budding bugologists with equal kindness. Who knows how it will affect them?

It is clear that you can’t necessarily convert entomophobic people to entomologists, but you may be able to provide some fuel for whatever pilot flame burns in youth that possess even a modicum of interest and curiosity regarding life on earth. Every now and again, you encounter some really exceptional students, and those are the individuals that can make the smallest gesture extremely important. I have been lucky to associate with a number of such exceptional undergraduate students here at UNBC. Some have continued in entomology while others have moved to other disciplines. Three former students who carried on to pursue doctoral degrees are noteworthy.  One is now a pollinator specialist with the Government of Alberta, one is a PhD student at York University, and one is a PhD student at Utah State. All have published in entomology journals, and are obviously successful. More importantly in the context of this blog, they have all kept in touch and expressed appreciation for the mentoring they received here at UNBC.

The author and his son looking at tidal pool life, Lighthouse Park, North Vancouver, BC, 1997 (Photo L.M. Friskie).

The author and his son looking at tidal pool life, Lighthouse Park, North Vancouver, BC, 1997 (Photo L.M. Friskie).

The ESC grants are relatively small, but they may make a huge difference in someone’s life and future career path. Therefore, as Chair of the Science Policy and Education Committee, I hope to be the recipient of several applications from regional affiliates this year.

Myrmica brevispinosa, the short-spined ant
, , ,

The trouble with common names

By Dr. Staffan Lindgren, University of Northern British Columbia

————————

When teaching Invertebrate zoology, entomology or forest entomology, I am regularly asked by students if they can use common names. Mostly this request is precipitated by the perceived difficulty of memorizing, let alone pronouncing, Latin names. I am fairly relaxed about these things, particularly with forestry students, who are quite unlikely to become entomologists no matter how you define that term.  It should be clarified that forest entomology is taught within a Disturbance Ecology and Forest Health course at my institution (UNBC), with diagnostics in half of a separate lab course. My stock answer is thus that they may use common names as long as the name clearly defines the species they are referring to.

Foresters are prone to colloquial terms, whether with respect to insects, trees or other organisms. For example, subalpine fir (Abies lasiocarpa) is called balsam by many, if not most foresters in BC, even though it is a distinct species from balsam fir (Abies balsamea) of eastern North America. Similarly, Pissodes strobi, the white pine weevil, is called spruce weevil (a legacy of the days when this weevil was considered three separate species, two of which primarily infest different spruce species in the west) or simply leader weevil.  The reason, supposedly, is that it is the wood quality that matters in terms of trees, and the type of damage with respect to insects. The consequences of being a bit loose with the taxonomy of a particular species may therefore seem fairly inconsequential in forestry.

Incidentally, our forestry students have even more to worry about when it comes to pathology, which they have to learn at the same time, as the same biological organism often has two completely different Latin names (including genera) depending on whether it is the sexual or asexual form (why this remains an accepted practice is beyond me), and they often do not have common names. Add the fact that fungal species seem to change name more often than I change vehicles (I was going to write ‘shirt’, but didn’t want to gross anyone out making you think that I wear the same shirt for years), and it becomes rather a nightmarish proposition for the poor students.

When it comes to entomology in general, however, common names are most commonly used in casual conversation, particularly with members of the public. For entomologists this is usually not a problem, but for non-entomologists it can be very confusing.  For example, colloquial use of ‘bug’ is pretty much anything that is small and crawls or flies around. Taxonomically it is quite specific (Hemiptera: Heteroptera). Ladybugs (Coleoptera: Coccinellidae) are perhaps the most recognizable insects to people in general, but they are clearly not bugs. Plant lice (Aphidoidea and Phylloxeroidea), bark lice (Psocoptera) and body lice (Phthiraptera) represent three vastly different taxonomic groups. In addition, if the non-louse groups above were to be correctly written to show that they are not Phthirapterans, there should be no space – however for these common names that principle is never applied as far as I can tell. It is to differentiate dragonflies, damselflies, stoneflies, mayflies, whiteflies etc. from the true flies. For example, a dragon fly, if there were such a thing (and probably there is somewhere – perhaps a decapitating fly (Phoridae) comes close enough to earn that epithet!) would be a dipteran, whereas a dragonfly is not. How is a non-entomologist supposed to know that (assuming that it is important to anyone except us entomophiles)? Then we can go on to more obvious misnomers such as ‘white ants’, which aren’t ants (Hymenoptera: Formicidae) at all, but termites (Isoptera).

Going back to forest entomology, one can have all kinds of fun with some common names, the origin of some could serve as fodder for endless speculation. For example, when discussing the problems with common names, I ask my students what they think a sequoia pitch moth (Synanthedon sequoiae)(Lepidoptera: Sesiidae) would attack. The correct answer is naturally “mostly lodgepole pine, but not sequoia”. Similarly, the Douglas-fir pitch moth (Synanthedon novaroensis) commonly breeds in lodgepole pine, but as far as I know not in Douglas-fir. I then go on to western spruce budworm, which as the name does not imply primarily attacks Douglas-fir.

Myrmica brevispinosa, the short-spined ant

Myrmica brevispinosa, the short-spined ant

Clearly one cannot expect members of the public to keep track of Latin names of insects, so common names are here to stay. I was interested to find in a book I recently purchased (Ellison et al. 2012) that the authors had invented common names for every species by essentially translating the Latin species epithet. That creates an interesting situation vis-à-vis the attempt of entomological societies to standardize common names (http://www.esc-sec.ca/ee/index.php/cndb; http://www.entsoc.org/common-names). Nevertheless, some ants simply retained their genus name, e.g., Harpagoxenus canadenis became “The Canadian Harpagoxenus” (not sure why, as they named the genus “The robber guest ants”), Formica hewitti became “Hewitt’s ant”,  Myrmica brevispinosa (the species in the photo accompanying this article) is called “The short-spined ant”, and perhaps my favourite Lasius subglaber was named “The somewhat hairy fuzzy ant”. Common names aren’t generally that innovative, but Latin names certainly can be.

Many years ago May Berenbaum (1993) wrote a column on this topic. If students would all read Dr. Berenbaum’s eminently humorous take on how insects get named, they would without a doubt get a new appreciation for both Latin names and their creators, and perhaps feel less trepidation about memorizing them. Then not only true blue entomologists would be tempted to buy a bumper sticker that read “Sona si Latine loqueris” (Honk if you speak Latin) (Unverified from http://www.latinsayings.info/).

Berenbaum, M. 1993. “Apis, Apis, Bobapis….”, American Entomologist 39: 133-134.

Ellison, A.M., N.J. Gotelli, E.J. Farnsworth, and G.D. Alpert. 2012. A field guide to the ants of New England. Yale University Press, New Haven and London, 398 pp.

Smoky Mountains photo by Staffan Lindgren
,

A Canadian Entomologist in Knoxville: Report on the ESA meeting in Knoxville, TN

Today’s post is by Dr. Staffan Lindgren, University of Northern British Columbia, and Second Vice-President of the Entomological Society of Canada.

————————————

I have just returned from the Entomological Society of America conference in Knoxville, Tennessee, and thought that this would make a suitable topic for my first ever blog. As I attended the ESC-ESA JAM the week before, it gives me a suitable reference point for evaluating the ESA meeting.

The ESA conference attracted ~3000 or so delegates, and consisted of numerous concurrent submitted sessions, over  100 symposia, and a sizeable poster session, with each set of 300+ posters available for viewing one day. In return for the $400 registration, you received a name tag, a program book, and some items encouraging you to attend the exhibition. If you were savvy and well connected, you could get some free snacks and drinks at one or several of the numerous university or ESA staff-specific mixers held more or less nightly (yes, I am savvy and relatively well connected). As a younger man, armed with enthusiasm, curiosity and at least a modicum of drive, I would identify all the talks relevant to my own research, and then run between sessions to catch them. Now I tend to pick a symposium or session and sit through it, as I find that I am more likely to get exposed to new and different ideas that way. When you have over 100 symposia over four days, however, it is near impossible to catch even a fraction of the sessions you wish to attend. I also like to browse the posters, rather than identify specific ones, but again – during a big meeting like ESA it is sometimes hard to get the time. It is extremely helpful when posters have 8×11 versions that you can take with you to read later, and you could scan QR codes with information on the posters (as well as sessions and exhibitors) into your smart phone/tablet if you were so inclined (I haven’t quite gotten there yet).  I would think these approaches are the future when the techies replace us old traditionalists. The meeting also had virtual posters for non-North American students unable to attend the meeting, which was a neat idea (even if I didn’t get around to looking at them either)!

Smoky Mountains photo by Staffan Lindgren

Smoky Mountains photo by Staffan Lindgren

The location of a meeting is obviously important. Knoxville has a lot to offer, not least of which is the “body farm”, or the University of Tennessee Forensic Anthropology Center Research Facility. Tours were offered to the site, but I don’t have the stomach for it, so I went with some colleagues to the Smoky Mountains instead (we did look at hemlock woolly adelgid, so it was an entomological trip!) The layout of the meeting venue is extremely important, I realized. Ideally, the session rooms should be organized around a central area, so that attendees have a chance to interact. The Knoxville meeting was held at the conference center, the layout of which was not conducive to personal interaction, unfortunately. It consisted of a three-storey square building, with the meeting rooms in the middle surrounded by a walkway, and the exhibition hall in the basement.  Consequently you could spend the entire meeting there and still not meet up with colleagues. I ran into two colleagues the evening of the last day I was there, and I failed to find one colleague I was actively looking for!

There are positives and negatives with every meeting, but when contrasting the Edmonton ESC meeting with the Knoxville ESA, or any other ESA meeting for that matter, I think ESC takes the prize both in terms of what you get for your registration fee, and ability to network with colleagues. The scientific program at the ESC meeting was of very high caliber as well, particularly the plenary session. Being small isn’t always a good thing, but when it comes to scientific conferences, I think it is a definite advantage. What do you think?