• G*Power: Statistical Power Analyses for Windows and Mac

http://www.gpower.hhu.de/

https://www.youtube.com/watch?v=5ccl4nmtUpM

  • Power and Sample Size

http://powerandsamplesize.com/

  • Sample size estimation and statistical power analyses

https://www.researchgate.net/publication/265399772_Sample_size_estimation_and_statistical_power_analyses

  • PASS Documentation

https://www.ncss.com/software/pass/pass-documentation/

  • NCSS PASS YouTube Channel

https://www.youtube.com/channel/UCNcvYpuMaaaFaAanqkFXRbw/featured

  • Tests for Two Means in a Repeated Measures Design

https://ncss-wpengine.netdna-ssl.com/wp-content/themes/ncss/pdf/Procedures/PASS/Tests_for_Two_Means_in_a_Repeated_Measures_Design.pdf

Power calculation for comparing sample means from two paired samples

https://www.youtube.com/watch?v=RCox1fE8rQw

  • JAMOVI ile orneklem sayisi belirleme

https://www.facebook.com/groups/sayisalarastirmayontemlerindesongelismeler/permalink/167090167301796/

http://www.bwgriffin.com/workshop/Sampling A Cohen tables.pdf

Karsimiza cikan ikinci secenek "minimally-interesting effect size" (ilgilenilen en dusuk etki buyuklugu) bu analizin en onemli bolumunu olusturmaktadir. Arastirma oncesi orneklem belirlemek istiyorsaniz nasil bir etki buyuklugu beklediginizi belirlemeniz gerekir. Etki buyuklugu iki grup arasinda ne kadar buyuk bir fark oldugunun bir olcusudur (Cohen, 1992'ye bakmanizi oneririm). Beklenen etki buyuklugunu belirlemenin en kolay yolu literaturdeki benzer calismalardaki etki buyukluklerine bakmaktir.

Bagimsiz iki grubun karsilastirilamsinda kullanilan en populer etki buyuklugu olcusu "Cohen's d"dir. 1 birim d iki grup arasinda 1 standart sapma fark olduguna isaret eder. Cohen (1992) d=0.2 kucuk, d=0.5 orta, ve d=0.8 buyuk etki buyuklukleri olarak onermistir.

"power by effect size" tablosu ise eger "gercek" etki buyuklugu tahminimizden farkliysa ne olabilecegini aciklamaktadir.

Ornegin gercek etki buyuklugu "orta derece" degil de daha kucuk ise (ornegin 0.3'ten kucuk - tablonun ilk satiri), grup basina 86 gozlemimizin istatistiki gucu %50 den daha az olacak ve calismamiz buyuk ihtimalle bu etkiyi tespit edemeyecek ("likely miss").

buyuk etkiler soz konusu oldugunda daha kucuk orneklemler yeterli olabilmektedir

  • Selecting a sample size for studies with repeated measures

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3734029/

https://glimmpse.samplesizeshop.org/

https://samplesizeshop.org/

https://homepage.divms.uiowa.edu/~rlenth/Power/

https://sites.google.com/site/optimaldesignsoftware/home

https://www.statsols.com/nquery

  • POWERLIB:SAS/IMLSoftware for Computing Power in Multivariate Linear Models

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4228969/

https://github.com/SampleSizeShop/POWERLIB

  • SPSS SamplePower

https://www-01.ibm.com/marketing/iwm/iwmdocs/tnd/data/web/en_US/trialprograms/U741655I36057W80.html

results matching ""

    No results matching ""