55
votes
The more important statistic: '90 percent of all women survived' or '90 percent of all those who survived were women'?
As they stand, neither one of Statement 1 or 2 is very useful. If 90% of passengers were women and 90% of people survived at random, then both statements would be true. The statements need to be ...
- 131k
30
votes
Recommended terminology for "statistically significant"
I don't think the objection is to just the term "statistically significant" but to the abuse of the whole concept of statistical significance testing and to the misinterpretation of results that are (...
- 128k
25
votes
What is the point of reporting descriptive statistics?
The point of providing descriptive statistics is to characterise your sample so that people in other centres or countries can assess whether your results generalise to their situation. So in your case ...
- 18.4k
19
votes
Report power if result is statistically significant
Context: I wrote this answer before the OP clarified that they are working with a large dataset, so the study (probably) has sufficient power. In my post I consider the more common case of a small ...
- 10.7k
18
votes
Accepted
Should I report non-significant results?
Yes, non-significant results are just as important as significant ones. If you are reporting any result, always include the df, test statistic, and p value. And in that case, you should state the ...
- 849
18
votes
The more important statistic: '90 percent of all women survived' or '90 percent of all those who survived were women'?
At its face, the conditional probability of surviving conditional on sex is more useful, simply because of the direction of information flow. A person's sex is known before her or his survival status,...
- 98.5k
17
votes
Are all 20 subjects the same height if the standard deviation of the sample is reported as 0.0?
According to this biology SE thread, the standard deviation of male adult height is about $0.07$ metres, and of females is about $0.06$ metres.
Rounding these to one decimal place would give $0.1$ ...
- 23.9k
11
votes
Accepted
Is it appropriate to report the mean, standard deviation and coefficient of variation for binary measurements?
It's overkill, but may be justified for presentation reasons.
Usually, the total number $N$ of measurements is known. If we then know the number $N_1$ of "1" entries, we can calculate all ...
- 131k
11
votes
How to report my log transformed (+1) data?
I am not quite so negative on $\log\ (y + 1)$ or more generally $\log \
(y + c)$ for some constant $c$ as some colleagues. For $y$ read also $x$ according to taste or circumstance.
But three points ...
- 59.5k
9
votes
Accepted
How do you report results from a Beta Regression (R output)?
The beta regression model can have two submodels: (1) a regression model for the mean - similar to a linear regression model or a binary regression model; (2) a regression model for the precision ...
- 16.1k
9
votes
Accepted
Justification for reporting non-significant effect sizes
A plausible answer:
While some researchers prefer hypothesis significance over effect sizes in reporting, it is often beneficial to complement coefficient estimates with effect-size measures, ...
- 2,373
9
votes
Recommended terminology for "statistically significant"
In my opinion one of more honest yet non-technical phrasing would be something like:
The obtained result is surprising/unexpected (p = 0.03) under the assumption of no mean difference between the ...
- 5,114
9
votes
Accepted
minimum total N (value for denominator) when calculating a percentage
For what purpose are you giving this list of percentages? What do you want your readers to get from it?
I know of no rule for this. For some purposes 50 might be a reasonable, but quite arbitrary, ...
- 57.6k
9
votes
Report power if result is statistically significant
Roughly speaking, observing a significant result in a test with low power means that the observed result is unlikely both under the null hypothesis and under the alternative. So interpreting such a ...
- 5,410
9
votes
What are the benefits of classical ANOVA F compared to Welch's F?
Several points:
The first thing is for the significance level calculations to be valid the assumption would be about the situation when $H_0$ is true. It would be possible for the distributions to ...
- 291k
8
votes
Accepted
Contradictory test results
This is indeed a problem (one of many) with hypothesis testing. This particular scenario is discussed in detail in the paper by Andrew Gelman and Hal Stern titled The Difference Between "...
- 8,767
8
votes
Accepted
Reporting results from a GLMER
Update (2024/08/10): I have added some additional points of consideration and citations to enhance the previous answer written here. Otherwise, the answer as a whole is essentially the same.
As Ben ...
- 18.5k
7
votes
Accepted
How to report a linear mixed effects model for those who are unfamiliar and skeptical?
I partly take side with the reviewer on this one. You are interested in the effect of your parameter of interest — given the rest of the model. It is hard to interpret the results and to check the the ...
- 1,188
7
votes
Should I report non-significant results?
If you are publishing a paper in the open literature, you should definitely report statistically insignificant results the same way you report statistical significant results. Otherwise you contribute ...
- 43.2k
7
votes
The more important statistic: '90 percent of all women survived' or '90 percent of all those who survived were women'?
The first indicates that saving women was probably of high priority (irrespective of whether saving men was)
The word "priority" comes from the Latin for "before". A priority is something one comes ...
- 5,321
7
votes
Recommended terminology for "statistically significant"
I agree with the answer by Peter Flom, but would like to add an additional point on the use of the term "significance" in statistical hypothesis testing. Most hypothesis tests of interest in ...
- 133k
7
votes
How to report a P-value?
When is it appropriate to give the exact P-value, instead of writing e.g. P<0.05 (also in case of non-significant P-values)?
As a general guideline you want to convey as much information about ...
- 86.5k
7
votes
Accepted
Good references on communicating the results of a statistical analysis to laypeople or non-expert stakeholders?
I picked up lots of useful tips from The Art of Statistics by David Spiegelhalter. I think he does an exceptional job of communicating some very abstract concepts without flattening the nuance in the ...
7
votes
Accepted
The "detectseparation" package. How to interpret its results?
As it says in the printout, these values indicate existence of the maximum likelihood estimates. 0 means they are finite (i.e. no complete separation), whereas infinite values are for completely ...
- 5,683
7
votes
How to report my log transformed (+1) data?
You wrote
Say that i have a variable with lots of 0 values that needs
log-transforming so I do log(variable+1) to transform it
No, I won't say this. Doing this log transform of + 1 is not a good ...
- 128k
7
votes
What are the benefits of classical ANOVA F compared to Welch's F?
A good reference that addresses your question, citing appropriate literature, is Delacre, M., et al. (2019). Taking Parametric Assumptions Seriously: Arguments for the Use of Welch’s F-test instead of ...
- 12.1k
7
votes
What descriptive statistics to report for a paired sample when using a nonparametric test?
Because your sample is paired, you should report the median and IQR of the difference between A and B. You can also report statistics about A and B separately, but the key test statistic will be ...
- 5,893
6
votes
Reporting non significant results
One of the things which makes subsequent synthesis of primary studies and their meta-analysis difficult is when the authors of the primary studies just say "not significant". If you are convinced that ...
- 18.4k
6
votes
Accepted
Report GLM and Posthoc with emmeans in APA format
I think in your paper you should show a table of the EMMs and SEs, and another table showing the comparisons among them and their Tukey-adjusted P values. Possibly you should do them with ...
- 21.3k
Only top scored, non community-wiki answers of a minimum length are eligible