GPower - Science topic

As sample size calculation is extremely important for RCT, are there any recommendations for simple tools? For example, can GPower be used for the calculation for RCT?

Dear community,

I can't seem to find any clear information on power analyses work so I'm throwing my question out onto the internet hoping for answers. :)

I am doing a within-subjects design with two independent factors. The dependent variable is binary correct/incorrect.

There are 18 test items.

9 items vs 9 items for Condition A (2 levels)

3 vs 3 vs 3 within each of those 9 for Condition B (3 levels)

I want to be able to detect significance for an Odds Ratio of 1.800, 2-tailed.

In a similar experiment, the lowest mean for a condition was .177.

My question is how I should use GPower to get a minimum sample size?

I've tried z tests -> logistic regression and it gives me a sample of 533, but there is no field to input the number of measures or independent variables, so I'm not sure how to interpret the result.

Or does it mean that I need that many individua

Am I dividing that number by 18 for the 18 test items? (approximately 30 people)

Am I dividing it by the number of conditions? (3x2=6 - approximately 90 people)l participants? (impossible)

Any help would be greatly appreciated!!

How to use GPower for unknown population ?

hope anyone can teach up .

Hello everyone,

I've recently submitted a manuscript, which was sent back to me with reviewers' comments and suggestions. In one of them, they asked me to perform a sensitivity power analysis given my actual sample size. My experimental design was a three-group intervention (i.e., one control and two experimental conditions) with a binary outcome. I've been struggling to define which family and statistical test I should choose for that purpose (I'm using GPower), for example: an F test family, in which I can add the number of groups and select an ANOVA fixed effects test, or a z test, in which I cannot include the number of groups but I it is the type of model I used to test my hypothesis.

I'm not sure if my reasoning is correct, but I'd much appreciate any help you can provide!

Hello everyone, does anyone know how to calculate the simple size for a 2(gender: male, female)*2(culture: Asian, European)*2(age: children, adult)*2(direction: back, front)*2(position: left, right)*2( condition: confort, non-comfort) repeated ANOVA? We have 6 factors, among these, gender and culture are between-factors, while age, direction, position, and condition are within-factors. I'd appreciate it if someone can help me.

I'm planning to use a cox regression in a future study exploring time-to-event or survival analysis comparing a control with an experimental group. I've seen sample size calculated through several packages, but I prefer G*Power and wanted to know if anyone's done this. Any resources would be appreciated.

Hi, I am looking for different approaches to calculate sample size for a repeated measures within subject design. What is the critique for using GPower and is it justified? And what are reliable alternatives? Is R really a good alternative or truly just doing the same as GPower?

Thank you!

Please could someone help me to calculate the sample size I would need for Moderation Analysis (Hayes' model 2) - I am interested in considering how attachment style (avoidant vs anxious) moderates the relationship between cognitive behavioural therapy and outcome measure (PHQ-9).

How do I work out sample size using G-Power?

Independant variable: Mode of delivery (online versus remote delivery)

Moderation variables: 1) Anxious attachment style and 2) Avoidant attachment style

Dependant variable: Score on Patient Health Questionnaire

There are no co-variates.

I am working on a study which involves moderated mediation analysis (1 IV, 1 Mod, 1 Med, 3 DVs), and I need to calculate the required sample size for my analysis. Is there any software or guideline I can follow for this task? Ideally, I wish to learn the principles so that I can apply for other situations as well. I have been using GPower but it does not have the option for moderated mediation analysis.

I have two experiments one with 3 participants as a group and the other with 4 participants as a group. How would I calculate the sample size if I would like to use HLM analysis? (medium effect size alpha 0.05 power 0.80) Thank you in advance!!

Does control group included when calculating the sample size using Gpower F test one way ANOVA? I have 3 study groups to compare with a control group (4th group). When calculating sample size using Gpower, the number of groups should be 3 or 4?

Good day Gurus!

I am new to research and which test to run as well as sample size.

I have three pre-tests (two scales and a writing prompt- gathering data from the writing), participants than randomized into three groups, separate intervention for each group, then three post-tests(writing prompt to test if there is a change from the intervention and same two scales from pre-test) with a survey at the end. I am evaluating the changes after the intervention, to hopefully see a positive change.

I have tried GPower, yet not sure how to pick the test which would give me sample size. I am thinking a multivariate comparison, yet that is why I am asking.

Independent Variables - Intervention

Dependent Variables - positive change, post-test scales, second writing prompt that should change due to intervention

Controls - Pre-test scales

Suggestions?

Thank you in advance. JDM

Please help: I need to calculate the g*power for my study. I have two predictor variables and one criterion variable. All measured as continuous scale.

Three research questions: Rq1-Does mindset predict GPA? Rq2-Does SES predict GPA? Rq3-Does mindset predict Gpa while moderating for SES?

I’m thinking that rq1&rq2 needs a simple linear regression/bivariate regression.

Im thinking rq3 needs a moderated multiple regression.

Will I run gpower once or twice? (For Two seperate test?)

What type of gpower is best for linear regression with two predictor variables and one dependent variable?

Hi everybody, I have a question about calculating the sample size with the software. There is various software such as GPower and NCSS. PASS in this area. which one is better? Can anyone guide me? For example; how can I work with NCSS. PASS software? Thanks

David Morse you helped me previously, please can I check with you my question is : the impact of working from home on loneliness, creativity and resilience. I know I will be using multivariate regression but what am I putting in G Power to work out the sample size. I am assuming possible control variables of 1)ethnicity, 2)education level, 3) type of work and my IV is working from home. My 3 DVs are loneliness, creativity and resilience. What is the best way to work out the sample size please?

Is a-priori analysis with an effect of f2 of 0.08 reasonable, and alpha of 0.05 and power of 0,95 ? which test is best? and how do I do this please? I think you mentioned a generic F Test or Manova

I am running different tests (t-tests, correlation, anova) and therefore have computed different power analyses. Should I report all of these, or only the test which needs the most participants to ensure enough power (as therefore all other tests will have enough power)? Thanks in advance!

Hey everyone!

I wonder what is the right way to calculate my sample size for two unrelated groups comparison if I am to base my effect size on the MDC and SD of my primary outcome.

In g power when going to the effect size calculation there is a place to enter the mean and SD of two groups. So what should I type there?

Thanks!

Hi. I am currently doing a research to compare between balanced and unbalanced bilinguals. I did a Gpower but I am not convinced whether that is the right amount or not as I have read previous literature, and the amount of the bilinguals in their research is not that big. Hence, I would really appreciate it if you could share with me any articles regarding this issue. Thank you so much.

Hi, I want to perform an a priori power analysis to determine the sample size for my study. I have one between subject variable with two levels (I assume number of groups = 2), three dependent variables (I assume response variables = 3). I am just interest in the main- and mediation effects. Alpha and Power are specified in the picture. Now, I was wondering about the effect-size categories (small, medium, large) because I cannot find anything about Pillai's V. What would be a medium effect size for f^2 (V)? Also, I thought that for MANOVAs, the partial eta-squared is indicating the effect size?

Best!

Hi everyone,

I tested an SEM model with 2 IV, 4 mediators and 1 DV on a sample of 1000 participants (see attached figure). Could you please help me to find an estimation for a good sample size using power analysis for this multiple-mediator model.

Best,

Robin

Hi all, I'm currently working on a research to finish my masters in Marketing. But I'm struggling with what test to run and how to estimate the sample size by using the GPower tool.

I'm investigating the following relationship: IV (2levels) --> DV (interval scale), and how a moderator (2levels) affects this relationship. I'm conducting a mixed design, the IV is measured within-subjects, and the moderator is measured between-subjects. There are also covariates added.

Do I need to run an ANCOVA or an ANOVA repeated measures?

Thanks a lot, in advance!

Best,

Fleur

Hi, everyone

In relation with the statistical power analysis, the relationship between effect size and sample size has crucial aspects, which bring me to a point that, I think, most of the time, this sample size decision makes me feel confusing. Let me ask something about it! I've been working on rodents, and as far as I know, a prior power analysis based on an effect size estimate is very useful in deciding of sample size. When it comes to experimental animal studies, providing the animal refinement is a must for researchers, therefore it would be highly anticipated for those researchers to reduce the number of animals for each group, just to a level which can give adequate precision for refraining from type-2 error. If effect size obtained from previous studies prior to your study, then it's much easier to estimate. However, most of the papers don't provide any useful information neither on means and standard deviations nor on effect sizes. Thus it makes it harder to make an estimate without a plot study. So, in my case, when taken into account the effect size which I've calculated using previous similar studies, sample size per group (4 groups, total sample size = 40 ) should be around 10 for statistical power (0.80). In this case, what do you suggest about the robustness of checking residuals or visual assessments using Q-Q plots or other approaches when the sample size is small (<10) ?

Kind regards,

Hello! I am wondering for the value of power to put as input parameters when computing the critical effect size via a sensitivity analysis on Gpower. Indeed, given that "typically, although not always, a critical effect size correspond to the effect size the study had 50% power to detect" (Lenth, 2007, cited by Lakens et al., 2018), should we put 50% in the power input estimates in senstitivity analysis (like we put 33% when using the small telescope approach) or Gpower calculates critical ES automatically with the "standard" 80%, 90% or 95% (or other) of power ? Thanks in advance

I am conducting an experiment with one within factor (8 levels), and will use a RM ANOVA to analyze the data. I would like to find the appropriate sample size using power, alpha and effect size from a previous study (eta² = 0.6).

How can I do this?

GPower is recommended a lot, but it tells me a sample size of only 2 subjects is enough.

Thank you!

Hi, I have a simple question.

I am hoping to perform a power analysis/sample size estimation for an RCT. We will be controlling for baseline symptoms, and using post-treatment or change scores as our outcome variable, Ie we will use an "ANCOVA" designs showed to increase power: https://trialsjournal.biomedcentral.com/articles/10.1186/s13063-019-3671-2

Would any body be able to point me towards the best tool for sample size estimation for such a model?

thanks!

GPower is only equipped to deal with calculations of bivariate regressions and needs SD s for both predictor samples.

However, if one has a multiple regression with more than 2 outcome variables (multivariate) , is there a way to calculate sample size with an alpha of 0.05 and assuming a moderate effect size of 0.35 apriori?

thanks in advance.

Hello, everyone. I have no idea about how to calculate the simple size if I used repeated- measures ANOVA with time (PRE vs POST) as the within-subject factor and group (young vs olderly) as the between-subjects factor. Am I going to use ANOVA:Repeated measures, within-between interaction? And I am not sure the number of measurements, is it four?

Hi, I am going to execute an experiment with the four categorical variables (A=2 levels, B=2 levels, C=2 levels, D= 5 levels) and one dependent continuous variable. There are total of 40 groups that I want to compare (2x2x2x5 = 40). For each group, I will hire separate participants, so it will be a between subject design. since I need to compare 40 different groups and any interaction effects, I need 4-way ANOVA.

In the GPower software, I selected the F test ANOVA (Fixed effects , special, main effects and interaction). See the Figure for more details. Now, my question is whether the total sample size is per group? Am I doing it right?

Hello, I am conducting an experiment for my dissertation and looking at the effect that relaxing photos have at mood , physical and mental health. There is only one group that is going to be used for this study (there is no control group). Participants will be asked questions on mood, physical and mental health (3 dependent varibles ) presented with the photos (Indepented variable) and then asked the same questions prior to the exposure to photos. The point is to see if the photos will have an effect or what effect on these 3.

The problem is I dont know how to best analyse the data in SPSS and the steps I need to do, as well as find the sample size I need for the study using Gpower because I dont know what statistical test I am using and how to find the effect size.

I want to explore if the amount of time (usage) spent on the Social Media IV1, with two levels , one is light and the other is heavy use, Iv2, body image with negative and positive levels, Iv3 self-esteem with levels of high and low, and my DV is sexual satisfaction. If these variables have an effect on sexual satisfaction and I need to determine my sample size! please send me if you can help me! SOS!

For my dissertation I am investigating help-seeking behaviour. Specifically, measuring help-seeking (actual and intended) using the GHSQ against different age groups, and also the effect of awareness of mental health campaigns (number of campaigns they are aware of). I am aware that gender may cause error variance, therefore have decided to measure gender also as a covariate. I am correct that to analyse this I would use a two-way ANCOVA? And how would I input this into GPower to calculate a sample size?

Thank you in advance!

Hey all,

I conducted an experiment for my bachelors thesis where I used ordinal logistic regression to analyze the data. My two independent variables are binary and my dependent variable is ordinal.

Now I wonder how to find out the minimum sample size using results from my low-scale pilot test.

I have tried using gpower but couldn't find a guide which seems to suit my situation online and therefore am a little lost right now. Can I use the same calculations as with a normal logistic regression?

Best,

Fabio

Hi,

I am currently doing a quantitative study (between-subjects design) to measure the degree to which alexithymia, emotional expressivity, mindfulness and coping strategies predict emotional eating, where emotional eating is the outcome variable and the rest are the predictor variables. The target population the study is done on is university students and black Asian or ethnic minority individuals (BAME). the measures i am using are: Toronto Alexithymia Scale; (c) Emotional Eating Scale; (d) the Mindful Attention Awareness Scale; (e) Berkeley Expressivity Scale; and (f) the Brief Cope Scale. All of these are likert scales.

How do i work out the sample size and actual power in GPower analysis* programme. any attachments tutorials to help me would be appreciated

Thanks

Hi, I have used the package pwr in R, but it falls short when compared with the GUI program G*Power (https://gpower.hhu.de/).

G*Power has a wealth of options and analyses but pwr is more basic.

Are there any other packages that incorporate all the features that G*Power has?

Greetings.

My study design is as below:

GPower computation details as below:

Kindly advise on the above to calculate the sample size eventually.

Thank you.

Say I'm using GPower to calculate the number of participants needed in a multiple regression to detect one of the predictor's unique effect. I want to be able to detect a correlation of .3. What you typically do is select:

- Test family: F tests

- Statistical test: Fixed model, R2 increase

- Type of power analysis: A priori (but I guess for my question, it really doesn't matter)

GPower ask for an effect size (f2), and has a tool to convert R2 to f2. In the toolbox, you can select a "Direct" input, which is partial R2. So from what I understand, GPower uses partial correlation has an input.

Is it strictly for partial correlation, or is it OK to use this for part correlation as well? I know both return the same p-value (it's just two different beta-to-correlation transformation), so I'm not even sure if it is relevant at all.

Thanks in advance for any insights on this question!

My study is purely on studying the effect of x on y, but I face difficulties in calculating the sample size of my social research using GPower. My question is which Test family and Statistical test should I use in Gpower.

P.S: I want to reject the null hypothesis and prove the alternative one (that there is an effect).

Gpower requires that df numerators be specified - can anyone advise how to estimate these in order to determine sample size? Thank you

I am about to start a randomized controlled study consisted of 3 groups. I want to know how I can use Gpower software to calculate the appropriate sample size for this study?

You  know we have not any Identified method for sampling in SEM method,so can we use sampling software for example Gpower ,NCSS,IBM sample size for our sampling .if yes ,HOW ?

Hello.

Recently I have become more familiar with determining à prior sample sizes, but mainly with T tests.

I will need to use those procedures for a F test (repeated mesures, within-between groups) and I am struggling a little. A couple questions that could help me:

1) In the typical "non-inferiority trial" is there a value of ES that is theorically atributed in this cases?

2) Yesterday I realized that Gpower has an calculator to help determine the ES. In the category "between groups" I could determine the ES to be 0.5

Under the same exact conditions, can I use that ES in "within-between" analysis or that would be an error?

The calculato for "between groups" allow to ES to be determined by means (that is fine with), while "within-between" requires variances and is beyond my stats skills. :)

Thanks.

Hi all,

In short, I am looking how to calculate the sample size based on a hierarchical regression.

Im conducting an experiment where I have 8 predictors in the main analysis. Step one includes 6 different scales (predictors) (IF) and step two includes two (CH).

I have two previous studies that have looked at the relationship between IF and the dependent variable, and three that have assessed CH and the dependent variable but none that have regressed both IF and CH on the DV. Also, the studies looking at IF and the DV have used less scales in their regression models than what I am using. I am therefore not sure if I can use effect sizes (f2 / R2 or F-value) reported by other studies since there are different amount of predictors? I was hence thinking of using the standardised coefficients (B) but.... how do I do that when I dont have B for all predictors? nor do I know how to combine all the coefficients to "one model" since it is a (hierarchical regression) in order to calculate sample size.

And yes, I know how to use all of the "typical" calculators, like, daniel sloper, GPower, Kelley & Maxwell etc.. Its the estimation and justification bit I struggle with.

My topic is in the HR field and i've 5 IVs, 1 mediator and 1 Moderator variable. I'm using Smart PLS and i should use the G-power analysis for calculating the sample size, but honestly i don't know why specifically the G-power since there are other techniques for calculating the sample size? moreover if anyone could help me in understanding how to use it, because i watched lots of videos and they use different statistical tests and each test provide different sample size and i dunno on what base to choose in order to calculate the sample size needed.

Dear All,

In my research I have 1IV,1DV and 1MV.

Indicator for IV -2 question, DV-9 questions and MV - 7 Questions.

Using G*power analysis, specifically for F-Test in Multiple Regression and PLS-SEM analysis, (effect size f² = 0.15 (medium); α = 0.05; power = 0.95; number of predictor = _).

1.How I should determine number of predictor?

2. It is number of predictor is number of IV in study?

Kindly assist me.

Thanks In advance.

Hi,

I am trying to compute the required sample size for my experiment. I have already conducted a similar experiment before, and I used a PAIRED-SAMPLES t-test to test the difference between mean 1 and 2. This time, the logic behind the experiment is similar, but I will be using an INDEPENDENT-SAMPLES t-test.

My question is: when I try to workout the required sample size on gpower, do I select paired-sample t-test and input all of the required information, get my required sample size and then say I'll need to use this number of people for the independent-samples t-test?

In other words - can I use data from a paired-samples t-test to workout the required sample size for an independent-samples t-test?

Thanks!

I have applied hierarchical cluster analysis with three variables (stress, constrained commitment and overtraining) in a sample of 45 burned out athletes. While examining the minimum sample size requirements, I found out that there is no specific power analysis for that (?). Thus, I applied the rule of thumb: 5*m² (m=number of variables). Is statsoft or Gpower programs or any else program suitable to examine sample size requirements for applying hierarchical cluster analysis?

I need to know is there is a minimun number of subject to run a cluster analysis. My sample has 546 subjects, y I have two test of five factors each one. A revierw ask me why I think that this number is enough, it would be better if I can do it quoting some research.

Hello

I currently have two dependent variables (with 4 measurements each) and two independent variables

My question is how to calculate the sample size using Gpower for multivariate multiple regression. since we only have multiple linear regression in Gpower

what statistical test to use? (t-test? F-test?)

otherwise is there another toll to do the calculations?

Thank you

Dear Researchers,

I am currently wondering how to calculate the required sample size for my study.

I have the following study design:

- 1 Independent variable (metric)

- 3 Dependent variables (metric).

I want to test, weither the IV has an effect on the 3 DVs. Therefore, I need to do a multivariate regression analysis (as far as I know).

For the sample size calculation I planned to use GPower, which however does not provide the test of multivariate regression. Does anybody has experience with sample size calculation for multivariate regression models?

Would be really helpful!

Thanks in advance, Lissy

Hi,

I'm trying to perform a post-hoc power analysis for a multinomial logistic regression with interaction terms, and I couldn't find any reference for it. The nearest answer was in the GPower manual, but it is made for binomial logistic regression and does not mention any information about how to handle the interaction terms.

Thanks in advance,

For a correlational study, I conducted an a priori power analysis based on a previous experiment with similar measures. Given this I was able to estimate my sample size, and went using that.

However, when submitting the paper, a reviewer mentioned that the sample was too sample for yielding stable estimates and cited Lakens & Evers, (2014).

I read recently (while not well explained) that you can also conduct/report a post hoc sensitivity power analysis to corroborate a significant finding in a correlation analysis. (using gpower)

Is this correct? Can someone elaborate on how to report this in APA format, and what I can actually say. The analysis does show that my r value is higher than the critical threshold for 80% power at .05.

Thanks in advance.

Study 1: A developmental psychologist wants to examine differences in nightly sleep duration among easy, moderate and difficult infants. he knows that the effect size f would be 22, and he wants to reach 80% power with error probability of .05.

Study 2: A psychologist wants to examine the association between self esteem and life satisfaction, using rosenberg self esteem scale and satisfaction with life scale. She knows that the correlation p H1 would be approximately .30(correlation p H0 would be 0), and she wants to reach 95% power with an error probability of .001.

Study 3: A relationship scientist examined the difference in psychological well-being between married individuals and single individuals. He collected data from 160 people(80 married, 80 single). The first group’s well being average was 4.27 and SD was 1.3. He wants to know what his achieved power is.

I haven't ever used Gpower, these are example studies and I just want to know the right way to use Gpower and answers to these would be of big help. Thank you.

Can anyone please advised me on calculating sample size using Gpower? I want to see the effect of an intervention( independent variable: 2groups) on mental health (continuous) of participants and I have considered baseline outcome measure as a covariate? How can I calculate the sample size in this case? Furthermore, my two secondary outcomes (both are in continuous scale) are also affected by intervention according to the previous literature. In such case, do I have to consider them as a covariate? I am confused on numerator df and don't know what to put in both of these instances?

Hi,

We're looking at the frequencies, forms, and functions of adjectives found in child directed speech in different types of corpora e.g. shared book reading and toy play. We'd like to conduct a power analysis to find out the minimum of data points required. I've previously used GPower to calculate sample sizes required to detect a particular effect size for experimental data, but I'm unsure about how to do this for corpus data, given that we're more interested in data points required rather than number of participants.

Any help would be much appreciated!

Thanks,

Jamie

If I use the GPower 3.0 to determine the sutable size of Sample to SEM . Variables are (4 independent and 2 mediators and one dependant variable ) then the number of factors should I put in the programe should be 4 Or 6?   

thanks

Can we determine the appropriate sample size for logistic regression using GPower even before we start to collect the data? If yes, how can we set the odd ratios (illustration in the figure)

Im trying to determine apriori my ideal sample size using the software G*Power. The model which I`m going to test is a moderation. Most of all I dont't understand what to do with the effectsizes. There are two effectsizes: one of my Regressor and one of my Moderator. Though there is only one area in G*Power in which I can add only one number.

Maybe there is somebody who worked with G*Power before and can help me with this?

I would like check the association BP response and few SNPs from a longitudinal study. I would like to which test is suitable to calculate the same in GPower software?