Determination of Minimum Sample Size Requirement for Multiple Linear Regression and Analysis of Covariance Based on Experimental and Non-experimental Studies

Abstract


 
 
 Background: MLR and ANCOVA are common statistical techniques and are used for both experimental and non-experimental studies. However, both types of study designs may require different basis of sample size requirement. Therefore, this study aims to proposed sample size guidelines for MLR and ANCOVA for both experimental and non-experimental studies.
 Methods: We estimated the minimum sample sizes required for MLR and ANCOVA by using Power and Sample Size software (PASS) based on the pre-specified values of alpha, power and effect size (R2). In addition, we also performed validation of the estimates using a real clinical data to evaluate how close the approximations of selected statistics which were derived from the samples were to the actual parameters in the targeted populations. All the coefficients, effect sizes and r-squared obtained from the sample were then compared with their respective parameters in the population.
 Results: Small minimum sample sizes required for performing both MLR and ANCOVA when r-squared is used as the effect size. However, the validation results based on an evaluation from a real-life dataset suggest that a minimum sample size of 300 or more is necessary to generate a close approximation of estimates with the parameters in the population.
 Conclusions: We proposed sample size calculation when r-squared is used as an effect size is more suitable for experimental studies. However, taking a larger sample size such as 300 or more is necessary for clinical survey that is conducted in a non-experimental manner.
 

Evaluation of the prevalence of illness anxiety disorder following the COVID‐19 pandemic in Iran and its related factors: a cross‐sectional study

“…According to other studies in this area, 20,21 the history of underlying illnesses like cancer, respiratory illnesses, diabetes, hypertension, and so on was one of the factors that predicted the development of illness anxiety disorder. This is because patients with underlying illnesses experience more fears and anxiety than other patients, which may increase their risk of developing illness anxiety disorder 22,23 . Because the results of the studies show the relationship among the illnesses like diabetes and high blood pressure with the severity of COVID‐19, so other underlying diseases such as kidney diseases, stroke, and cancers may show the same connection; however, proving this needs further investigation 24,25 …”

“…This is because patients with underlying illnesses experience more fears and anxiety than other patients, which may increase their risk of developing illness anxiety disorder. 22,23 Because the results of the studies show the relationship among the illnesses like diabetes and high blood pressure with the severity of COVID-19, so other underlying diseases such as kidney diseases, stroke, and cancers may show the same connection; however, proving this needs further investigation. 24,25 Another remarkable finding in this study is that in general, those who had a relative with a history of COVID-19 had a higher chance for illness anxiety disorder, but this connection in people who worked or studied in nonmedical fields, is several times more than those who are busy in medical fields; in confirming it, one can point to the results of another study which reports that the illness anxiety disorder score was higher in people with less work experience.…”

The christian religious education teachers’ attitudes toward the five-stage lesson plan framework in Kenya: A gender-based analysis

“…Multiple regression was also preferred over other models because it assesses the four explanatory variables simultaneously instead of separately [ [ 82 , 83 ]; [ [ 84 , 85 ]]. Multiple regression modeling could analyze these four explanatory variables [ [ 86 ]; [ [ [87] , [88] , [89] ]].…”

The Improvement of Generation Z Financial Well-Being in Pekanbaru

“…The population was more than 800,000 peoples (generation Z of 18-25 years). According to (Bujang, Sa'at, & Sidik, 2017), the minimum number of samples for the six variables tested using multiple regression analysis was 130. From 263 questionnaires, researchers only took 239 samples using purposive sampling and convenience sampling methods.…”

“…Multiple linear regression has ordinarily been utilized as a statistical tool to predict the dependent variables based on a set of predictors (Bujang et al, 2017).…”