The Stack Overflow podcast is back! Listen to an interview with our new CEO.
21

First, let's consider the original Wedekind et al (1995) study. Sample sizes seem reasonable, if the effect is not very noisy: [...] 49 female students (average age: 25.2 years, s.d. = ­4.0) and 44 male students (average age: 24.7 years, s.d. = 2.6) [...] ... and the statistical design accounts for individual differences in a fairly robust way: [...] ...


19

You've already gotten a decent answer to this, but I'll provide my own thoughts on the subject. Yes It's necessary. It is absolutely something you should do before beginning an experiment, and preferably something you should do in collaboration with the person who is going to be helping you analyze your data. To address a couple points: You'll see all ...


18

There’s an XKCD comic which explains the problem. Unfortunately, that comic is too big to post here. Briefly, a p-value of 0.1 says (roughly) that there’s a 10% chance (0.1) of the observed result being as extreme1 as it is simply due to chance (sampling variation from a population), assuming the null hypothesis is true. Often, 5% is more or less ...


16

Epidemiological studies analyze human observational cohort data to try to statistically link disease risk and biomarkers. For instance, it is know a well know fact that smoking increases the risk of lung cancer, yet before the 1950s it was widely considered otherwise until observational evidence proved irrefutable (Levin, 1950). Confounding However, ...


15

For very small systems like the worm c.elegans it must be possible to record from all neurons at the same time, at least optically. While it is true that whole organism optical recording is technically possible in C. elegans, I'm not aware of any published work where all neurons were identified and recorded from simultaneously and then combined with ...


11

In order to calculate power, you need to know the variance of the data being collected. You can only estimate this prior to actually gathering the data itself, so any a priori power calculation is itself just an estimate. This is why sometimes you will see small studies being conducted as pilot studies (for example, see (1)), which enable variance to be ...


11

kmm's answer is correct; I just want to add some of my points on what kind of data should follow Gaussian distribution. Unless you know from observation that a process doesn't follow a Gaussian distribution (e.g., Poisson, binomial, etc.), then it probably does at least well enough for statistical purposes. I won't fault kmm for this statement ...


10

It is correct if (and only if) albinism and Down syndrome are independent events, that is, if albino people is not more nor less likely to have Down syndrome than non albino people (and vice-versa). However, doing a quick Google search I can't find any source relating albinism and Down syndrome, and therefore independence seems a reasonable assumption ...


9

One thing that's certain is that the activity of bees varies according to time of day, so more important than how long you record for is probably at what time you record. If you always record at the same time of day, this should allow a reasonable comparison between results of different days. For example, recording from time 14:30 to 14:35 every day could be ...


9

Mechanistic model answers the how question. These models are usually biophysically detailed, and designed to be causal. Say you discovered a linear relation between blood pressure drug and heart rate. This would be a statistical model. It doesn't tell you how the two are related biophysically. One could build a detailed model that describes intermediate ...


9

You raise two issues, both of which might be better suited for stats.SE, but I think the questions are suitably biological to warrant an answer here. Do most biological processes follow a Gaussian distribution? Unless you know from observation that a process doesn't follow a Gaussian distribution (e.g., Poisson, binomial, etc.), then it probably does at ...


8

To answer this question in its entirety we have to split it into two questions: What are the underlying mechanisms of carcinogenity? One of the main mechanism behind carcinogenity is the mutagenity of the cancerogens, i.e. the ability to cause mutations, that are abberations of the cell DNA leading to uncontrolled proliferation. This classical paper ...


8

I understand this in the following way: For each probe you have two sets of measurements, one for ER+ and one for ER-. What you do is a T-test (to my understanding is that the "parametric" just emphasizes that T-test is a parametric test) on these two sets, testing if their mean is significantly different (they refer to this as "separated"). You repeat this ...


8

The NCBI Taxonomy statistics page displays the following information: There are currently 73540 genera, 331418 species, and 23127 taxa of higher order. Since the number of taxa decreases with the genericity of the taxon, there are probably around 20000 families, give or take a few thousand.


8

(World Health Organisation, 2014. p. 2) shows the adverse effect (side-effect) rates for the MMR vaccine. For instance, fever between the 7th to 12th day after vaccination occurs in 5 to 15% of vaccine recipients and rashes occur in 2-5% of vaccines, etc. Page 6 summarises the results in a table. As you point out, the adverse effects are variable in ...


7

Assumptions: Blonde hair is Homozygous Recessive and that the traits are strictly Mendelian. The parental generation must be both heterozygotes as at least one child is Blonde (bb). So your cross is Bb x Bb. Your square is going to look like this: _B_ _b_ _B_ BB Bb _b_ bB bb So of the question ...


7

if you are using NMR structures you might be making the mistake of using several superimposed structures - it would be nice to develop with x-ray structures at less than 2.0 A resolution for starters. Some of the models at low resolution can be sloppy, but submitted after 1994 or so will not have any center to center distances as that's when the x-ray ...


7

Due to my own woeful ignorance on the subject, I have been reading up on statistical methods recently. From what (little) I understand, the real answer to this question is: Yes, but only if you are doing Neyman-Pearson hypothesis testing and Absolutely not, if you are using Fisher p-values That is, the question isn't formulated correctly, because power ...


7

There don't appear to be many robust studies on this, but this study by De Longe et al in 2004 suggests that, whilst the volume of the ejaculate is increased after 5-days of abstinence (in most participants), for many this did not increase further after 8-days [1]. Here are the 11 participant included (along the x-axis), and for each participant there are ...


7

I'm assuming you're asking how to estimate population size based on numbers of individuals you capture. Yes, there are many approaches to this. The R library SPECIES-R offers a number of methods of calculating this. They are described in the manuscript SPECIES: An R Package for Species Richness Estimation. Some of the methods described are: Chao, A. (1984), ...


6

The Plant List has 642 families listed: http://www.theplantlist.org/1.1/about/#changes. For a quick comparison, Wikipedia lists 522 fish families: en.wikipedia.org/wiki/List_of_fish_families 136 mammal families: en.wikipedia.org/wiki/Mammal_classification 61 amphibian families: en.wikipedia.org/wiki/List_of_amphibians 57 reptile families: en.wikipedia.org/...


6

The 2011 paper How Many Species Are There on Earth and in the Ocean? indirectly answers this question as well as any other source you'll find I imagine. It estimates how many species there are total based on the rate of discovery of higher taxa; it includes plots of number of taxa over time for the major groups of life in Figure S1. Which gives: Animalia - ...


6

To choose the right statistical method (it is more than just saying "use the t-test") you need to think about your experiment. A good starting point is this figure from Bitesizebio: There are two relevant articles on that website: Let’s Talk About Stats: Comparing Two Sets of Data Let’s Talk About Stats: Comparing Multiple Datasets Probably also ...


6

To me, there are two issues that are mixed up here (if I understand you correctly). First, do you want to estimate the mean and variance for a statistical population (i.e. to characterize a larger population by independent samples), or do you want to calculate the actual density for a particular area, where you have counted all occurences in that entire area ...


6

@bshane has already provided an excellent answer to the question. However, I just found a recent review published in Molecular Ecology (Winternitz et al., 2016) in which the authors performed a phylogenetically controlled meta-analysis on MHC-dependent patterns of mate selection in humans and non-human primates. They used studies from seven primate species ...


5

I'm not an expert on Shannon-Weaver Index, but according to wikipedia it is the same as exponentially transformed Renyi entropy. If it is the case, you can compare them since they are scale invariant summary statistics. If you want error bars, you can always try resampling methods such as bootstrapping. Hypothesis testing can also be done with bootstrapping, ...


5

I guess you meant the population size stability. It is considered that the biosystems will increase their capacity of adaptation when evolving in very fluctuating environments. I believe the population stability is embedded in the adaptability of individuals. There is a measurement about it, evolvability, when the environment changes, the faster the ...


5

here I found an answer. Not sure how accurate is is, though. Data seems to be from the US Here is a rewrite of parts of that article: 266 days before birth, all we have is a fertilized egg. (~33% chance of living birth). The odds are calculated using data from in-vitro fertilization, and the next stage is 66%. In in-vitro, some 50% of the eggs cannot fix ...


5

In GWA studies you tend to analyze your "lead" SNPs in regions where genotypes are correlated (known as linkage disequilibrium). If you find an association between another SNP with the outcome, and this SNP is correlated with the original variant, you can perform a conditional analysis where you adjust for the original SNP in the model. This is to test if ...


5

I don't know about other groups, but about plants, number of families depends on the system you follow. Recent version of The Plant List (1.1) estimates about 352 000 species of Angiosperms and lists over 400 families. See http://www.theplantlist.org/1.1/browse/A/ It is very good and reliable source of information. Second very good source about plant ...


Only top voted, non community-wiki answers of a minimum length are eligible