Questions tagged [statistical-power]

Is a property of a hypothesis testing method: the probability of rejecting the null hypothesis given that it is false, i.e. the probability of not making a type II error. The power of a test depends on sample size, effect size, and the significance ($\alpha$) level of the test.

Filter by
Sorted by
Tagged with
-1
votes
0answers
9 views

Sample size calculation for experiment with two treatments [duplicate]

For an experiment I have to calculate the sample size. We want to observe how beliefs affect dishonesty and than add a treatment to see whether beliefs still have an effect when social image concerns ...
0
votes
2answers
37 views

Calculating sample size with power analysis

I have to calculate the sample size for an experimental study and took some sample distributions data from a previous similar research and calculated the means and standard variations for before and ...
1
vote
0answers
29 views

Determine distribution of a new test statistic

Question from my course project: Suppose we are conducting a two-arm trial. For the control group, there are $n_c$ patients, and their outcome is denoted by X, which has distribution $N(\mu_{c},\sigma^...
1
vote
1answer
16 views

Sample size calculation for linear mixed-effect models without data

I would like to know how to do a sample size calculation without data/simulation for mixed-effect models. To briefly explain my research, I want to investigate which method (A or B) is more effective ...
0
votes
0answers
36 views

Distribution of Product of Two Binomial Variables

Related to Probability mass function of product of two binomial variables I am doing a Power Analysis for Sample Size determination. I have two binomial distributions: ...
1
vote
1answer
25 views

Power analysis/sample size determination by simulation: Optimal algorithm for finding $N$

I want to calculate the needed sample size $N$ to achieve $80\%$ power for a complex longitudinal study design. Because of the complexity of the design and because I want to have control over every ...
2
votes
1answer
23 views

Ansari-Bradley Test Sensitivity to Median Differences: Should we subtract the median from each group?

The Ansari-Bradley test appears to be ineffective at detecting scale differences when the two distributions have markedly different medians. However, when I subtract the median of each group, I get ...
0
votes
0answers
6 views

Generate binomial data set with random effects for few observations per participant

I have a question regarding the generation of synthetic data for sample size estimation, in particular random effects, in a binomial data set with a small number of observations per participant. First ...
0
votes
0answers
21 views

Regression analysis with small but complete dataset (fully representing reality)?

I have a repeated measures design with about 16 cases and 5-6 points of measuring. Sometimes, 1-4 full cases or some points of measure are missing. (The measures are 20 numerical and categorical data ...
1
vote
0answers
12 views

G*Power settings to find minimum sample size for a 2x2x3 design

For a power analysis, does "number of measurements" refer to the number of LEVELS or the total number of CONDITIONS? I have a three-way (2x3x2) ANOVA that is entirely within-subjects, and I'...
1
vote
0answers
23 views

Power, split plot designs, and mixed models

My teammate and I are trying to determine the best way to allocated limited resources (shade cloth and fencing) in an experiment. We are considering 2 factors: light availability (2 levels) and water ...
0
votes
1answer
25 views

Simulating covariates for a cox-ph model power analysis

I have a dataset where I have the outcome data, and need to do a power analysis based on a missing covariate. E.g.: Surv(time, event) ~ X1 + X2 + ... + Xn + Z I ...
1
vote
1answer
27 views

Insufficient power from pwr.p.test

Using the pwr package, I determined the optimal sample size of 38. However, when I ran simulations using this sample size, I only get a power of 70%. What's going ...
0
votes
0answers
9 views

power, correlations and difference scores

Would it make a difference for a statistical power calculation whether the correlations are obtained using the simple measures or whether they are difference scores? Would you need more participants ...
1
vote
0answers
48 views

Statistical Implications of Beta(0,0) in Power Analysis

My team uses the code below to determine whether certain test rates of components will allow us to detect any significant degradation in component reliability. It compares the beta distribution for a ...
0
votes
0answers
14 views

Sample size for binary outcome and continuous (%) predictor

I have a project in which I have 60-70 samples. Outcome = binary (positive or negative case) Exposure = % of the time participants followed new guidelines (0-100%) Question - would it be correct to ...
1
vote
1answer
61 views

Post-hoc Power analysis to question a non-significant difference?

so we did this study about surgical success-rates in two groups (n=84 respectively). The outcome is binary (success vs. failure). A priori we expected them not to differ and, indeed, there was no ...
0
votes
0answers
33 views

Why the power of t-test increases with sample size?

Consider a normal sample $X_1,\ldots,X_n\stackrel{{\rm iid}}{\sim}N(\mu,\sigma^2)$ where both parameters are unknown. The $\alpha$-level standard t-test for $H_0:\mu=\mu_0$ versus $H_1:\mu\neq\mu_0$ ...
0
votes
1answer
15 views

How to determine sample size for mixed effects design

I'm designing a study with one within subject factor (intervention 1: baseline vs increased), a between subjects factor (intervention 2: present vs absent), and another between-subjects factor (...
1
vote
0answers
30 views

Power analysis for multinomial logistic regression

I am planning a regression analysis where a continuous independent variable predicts 3 categorical outcomes of a dependent variable. I believe this is done using multinomial logistic regression. ...
2
votes
0answers
18 views

Power analysis for Spearman correlation in R

I am wondering if the function pwr.r.test() in R can be used for all correlation coefficients. Does it matter if I add Spearman, Kendall or Pearson coefficient?
1
vote
0answers
14 views

Strong power for a treatment size within the 95% confidence interval?

I have a set of country-specific RD regression results showing very wide confidence intervals for my point estimates. The regressione are conducted through the local polynomial approach described by ...
1
vote
1answer
41 views

DF When calculating Chi Square Sample Size

I was wondering if someone could provide me some insights on sample size calculation for Chi Square tests. $$\begin{array}{c|c|c|} & \text{Group 1} & \text{Group 2}& \text{χ2} \\ \ \text{...
0
votes
0answers
5 views

How to compute cohort study sample size for proportion metric and continuous metric?

I'm trying to design an retrospective cohort study to investigate the effect of smoking on lung cancer, and I've built an propensity score matching model that can match users, shown as follows. $$ P(...
0
votes
0answers
33 views

Can the null hypothesis be **more** true than the alternative hypothesis or vice versa? [duplicate]

As the title says, my question is can the null hypothesis ($H_0$) be more true than the alternative hypothesis ($H_1$) or vice versa? For example, if the statistical power is really high (extremely ...
5
votes
2answers
56 views

After the sample size calculation what is the next step in clinical trials?

Suppose I want to test if drug A is effective to reduce the levels of some disease. From previous studies, the drug is estimated to reduce the disease by 0.5 units (mean difference) with a standard ...
1
vote
1answer
37 views

Sample size calculation with different standard deviations

when calculating the required sample size with a given power and p threshold, usually the calculator only requires two estimated population mean and standard deviation. is it always valid to assume ...
0
votes
0answers
9 views

Calculate power of the Kruskal-walis test [duplicate]

I think this has been asked before on here but it didn't get a straight answer. I would prefer to do it in R if possible.
2
votes
1answer
60 views

Minimum sample size and power test

Suppose I have access to data on the energy consumption of a city with a population of 30000 persons. Imagine I want to test a new technology of smart energy, i.e if the energy consumption decreases ...
2
votes
1answer
25 views

Ensemble vs statistical power

In regression modelling, I've seen two schools of thought: ensemble model vs. focus on statistical power (using one model). Proponents of ensemble models (i.e., bagging) argue that: Suppose $\...
0
votes
0answers
9 views

One-sided ttests in regression analysis with a clear prior

This question has been asked virtually the same in this post: Regression Slope and a Bilateral Test However, I still don't get my head around why when I have a clear hypothesis of a positive ...
0
votes
1answer
19 views

How to correctly interpret the results of Power Analysis?

Attached is the result of Power Analysis comparing the means of two independent samples. I am having hard time interpreting the "interpretations" as it is quite convoluted. Does it say the ...
1
vote
1answer
45 views

Calculate power of 2 sided Z A/B test - pooled version

I want to calculate the statistical power of a 2 sided A/B test where I have 2 groups of coin tosses (that is, from 2 different coins) and their respective successes. I want to test whether the two ...
34
votes
2answers
3k views

Which statistical model is being used in the Pfizer study design for vaccine efficacy?

I know there's a similar question here: How to calculate 95% CI of vaccine with 90% efficacy? but it doesn't have an answer, at the moment. Also, my question is different: the other question asks how ...
1
vote
0answers
11 views

Power Analysis for Time Series Models (ARIMA and neural network)

In short, I am building ARIMA (AutoRegressive Integrated Moving Average) and neural network time series (to be more specific, I am using the nnetar() function in R) models to make regression ...
4
votes
2answers
111 views

Calculate the power of a paired t-test in R manually

As an exercise I wanted to perform a paired t-test manually in R to refresh a lecture I had in the past. Everything went well, but then I thought about calculating the power of this paired t-test and ...
0
votes
0answers
17 views

Linear mixed model: Does the effect size of an interaction depend on the size of the main effects when size of interaction is kept constant?

Please also see my question here where I describe how I stumbled upon this problem and where I show the respective R-Code. I am running a power simulation where I specify the fixed effects for each ...
1
vote
0answers
35 views

How to interpret effect sizes in simr power simulation for mixed models?

I am conducting a power analysis for linear mixed models. I want to know what sample size is needed to find a specific effect size for an interaction. I am testing the interaction effect by comparing ...
3
votes
1answer
60 views

Power simulation (a priori) for mixed models when no prior data is available

The recommended way to do a priori power calculations for linear mixed models is to simulate data and then estimate power using the simr package, for example. All these approaches require you to ...
3
votes
1answer
116 views

How do I plot the power of the likelihood ratio test [closed]

I have a random sample of $25$ observations, which follow a normal distribution with mean $\mu$ and variance $100$. I have to graph the power of the likelihood ratio test of $H_0: \mu =0 $ and $H_A:\...
2
votes
1answer
74 views

Comparing 3 independent proportions (categorical)

I want to compare the proportion of correct answers. My variables are categorical (group condition type and answer) and my answer options are dichotomic (one trial, one question - correct/ wrong). ...
3
votes
1answer
28 views

Implications for sample size of multilevel/mixed-effects model when level 1 units are participant decisions and level 2 units are participants

I am planning a study that will use a hierarchical design. The study has participants engage in a decision-making task, wherein they make 40-50 decisions between 'Choice A' and 'Choice B' (I won't ...
2
votes
2answers
28 views

Estimating sample size required from previous study

Suppose I have a two phase experiment. The goal of the experiment will be to test if there are differences in proportions between two treatments. In phase one, I have no idea how many samples I will ...
0
votes
0answers
22 views

Number of replicates given effect and power and an experiment dataset

I came across this question in one of the modules on linear regression that I'm self-studying: Suppose an experimenter has decided on the settings for a two-variable problem, and these are the same ...
0
votes
0answers
11 views

Calculating power for a simple mediation model - Can I enter values observed from my own dataset into a Monte Carlo simulation covariance matrix?

I am going to conduct secondary data analysis. I need to figure out if the dataset has enough power for a simple mediation model. Each variable is observed and continuous from a cross-sectional study....
7
votes
1answer
46 views

3 Treatment Agronomic Experiment: Latin Square or Randomized Complete Block Design with 4 replicates?

I need help with designing an experiment. Suppose I am testing 2 fertility programs against a control in Lettuce, for a total of 3 groups (Program A, Program B, and Untreated Control). Let's assume ...
2
votes
1answer
73 views

Is the squared Cohen's $f$ the same as Cohen's $f^2$?

Question Is the squared version of Cohen's $f$ the same as Cohen's $f^2$? Or do they only share the same letter by coincidence? Explanation At first, I wondered if one can simply square Cohen's $f$ to ...
3
votes
4answers
78 views

Power test gives power=0.995

I am running a power test for an experiment where I have 122 controls (n1) and 184 experimental sets (n2). I selected a medium effect size of 0.5. I used the pwr.2p2n.test function in the pwr package, ...
2
votes
1answer
27 views

post hoc power analysis for SEM model?

What are my options for conducting any sort of power analysis post hoc? And if not a post hoc power analysis, what should I report?
0
votes
0answers
25 views

discrepancy calculating power of 1 sided 1 sample proportion test

I am running the R code below to compare two methods for finding the power for a one-sided (greater) test of a proportion versus a fixed value ie lower bound (ie one sample test) - I have sample size ...

1
2 3 4 5
24