Using GPower software to determine the sample size from the pilot study

Sultan Refa D Alotaibi, Dmitri Roussinov

Research output: Contribution to conferencePoster

Abstract

Recently, the estimation of the sample size is very important topic for the researchers, so it is an indispensable process for obtaining good results. There are two way to determine the sample size which are from literature review and from the pilot study. In this poster, we will focus on power analysis to get suitable sample size for our research. This power analysis is conducted to assess the minimum sample size requirement. GPower software application is used to perform a priori power analysis for the study taking multiple regression analysis for the validation of the measurement model. The required power was set at 1- β = 0.80. Level of significance was kept at α = 0.05. Effect size was kept at small range value of 0.02. For the measurement model each proposed latent construct has four indicators and the number of predictors is taken as 7. Using these settings, power curve was extracted. Figure 1 gives the power curve for different values of power ranging from 0.5 to 0.95. Power analysis clearly indicates that to achieve, power of 0.80, a total sample size of n = 725 is needed for this study. Therefore using the literature and thumb rule method and results of power analysis indicate that the total sample size needed for the study is around n = 750. Therefore, 800 respondents are taken as the optimal sample size for the investigation.

Fingerprint

Application programs
Regression analysis

Keywords

  • GPower software
  • sample sizes
  • priori power analysis

Cite this

Alotaibi, S. R. D., & Roussinov, D. (2016). Using GPower software to determine the sample size from the pilot study. Poster session presented at The 9th Saudi Students Conference, Birmingham , United Kingdom.
Alotaibi, Sultan Refa D ; Roussinov, Dmitri. / Using GPower software to determine the sample size from the pilot study. Poster session presented at The 9th Saudi Students Conference, Birmingham , United Kingdom.
@conference{ab33cbdf6b2145009fc8907479be0ddb,
title = "Using GPower software to determine the sample size from the pilot study",
abstract = "Recently, the estimation of the sample size is very important topic for the researchers, so it is an indispensable process for obtaining good results. There are two way to determine the sample size which are from literature review and from the pilot study. In this poster, we will focus on power analysis to get suitable sample size for our research. This power analysis is conducted to assess the minimum sample size requirement. GPower software application is used to perform a priori power analysis for the study taking multiple regression analysis for the validation of the measurement model. The required power was set at 1- β = 0.80. Level of significance was kept at α = 0.05. Effect size was kept at small range value of 0.02. For the measurement model each proposed latent construct has four indicators and the number of predictors is taken as 7. Using these settings, power curve was extracted. Figure 1 gives the power curve for different values of power ranging from 0.5 to 0.95. Power analysis clearly indicates that to achieve, power of 0.80, a total sample size of n = 725 is needed for this study. Therefore using the literature and thumb rule method and results of power analysis indicate that the total sample size needed for the study is around n = 750. Therefore, 800 respondents are taken as the optimal sample size for the investigation.",
keywords = "GPower software, sample sizes, priori power analysis",
author = "Alotaibi, {Sultan Refa D} and Dmitri Roussinov",
year = "2016",
month = "2",
day = "14",
language = "English",
note = "The 9th Saudi Students Conference ; Conference date: 13-02-2016",
url = "http://uksacb.org/ssc9/conference/poster-presentations/22-general/320-posters-engineering-and-computer-science",

}

Alotaibi, SRD & Roussinov, D 2016, 'Using GPower software to determine the sample size from the pilot study' The 9th Saudi Students Conference, Birmingham , United Kingdom, 13/02/16, .

Using GPower software to determine the sample size from the pilot study. / Alotaibi, Sultan Refa D; Roussinov, Dmitri.

2016. Poster session presented at The 9th Saudi Students Conference, Birmingham , United Kingdom.

Research output: Contribution to conferencePoster

TY - CONF

T1 - Using GPower software to determine the sample size from the pilot study

AU - Alotaibi, Sultan Refa D

AU - Roussinov, Dmitri

PY - 2016/2/14

Y1 - 2016/2/14

N2 - Recently, the estimation of the sample size is very important topic for the researchers, so it is an indispensable process for obtaining good results. There are two way to determine the sample size which are from literature review and from the pilot study. In this poster, we will focus on power analysis to get suitable sample size for our research. This power analysis is conducted to assess the minimum sample size requirement. GPower software application is used to perform a priori power analysis for the study taking multiple regression analysis for the validation of the measurement model. The required power was set at 1- β = 0.80. Level of significance was kept at α = 0.05. Effect size was kept at small range value of 0.02. For the measurement model each proposed latent construct has four indicators and the number of predictors is taken as 7. Using these settings, power curve was extracted. Figure 1 gives the power curve for different values of power ranging from 0.5 to 0.95. Power analysis clearly indicates that to achieve, power of 0.80, a total sample size of n = 725 is needed for this study. Therefore using the literature and thumb rule method and results of power analysis indicate that the total sample size needed for the study is around n = 750. Therefore, 800 respondents are taken as the optimal sample size for the investigation.

AB - Recently, the estimation of the sample size is very important topic for the researchers, so it is an indispensable process for obtaining good results. There are two way to determine the sample size which are from literature review and from the pilot study. In this poster, we will focus on power analysis to get suitable sample size for our research. This power analysis is conducted to assess the minimum sample size requirement. GPower software application is used to perform a priori power analysis for the study taking multiple regression analysis for the validation of the measurement model. The required power was set at 1- β = 0.80. Level of significance was kept at α = 0.05. Effect size was kept at small range value of 0.02. For the measurement model each proposed latent construct has four indicators and the number of predictors is taken as 7. Using these settings, power curve was extracted. Figure 1 gives the power curve for different values of power ranging from 0.5 to 0.95. Power analysis clearly indicates that to achieve, power of 0.80, a total sample size of n = 725 is needed for this study. Therefore using the literature and thumb rule method and results of power analysis indicate that the total sample size needed for the study is around n = 750. Therefore, 800 respondents are taken as the optimal sample size for the investigation.

KW - GPower software

KW - sample sizes

KW - priori power analysis

UR - http://uksacb.org/ssc9/

UR - http://uksacb.org/ssc9/conference/poster-presentations/22-general/320-posters-engineering-and-computer-science

M3 - Poster

ER -

Alotaibi SRD, Roussinov D. Using GPower software to determine the sample size from the pilot study. 2016. Poster session presented at The 9th Saudi Students Conference, Birmingham , United Kingdom.