Question
Asked 11th Sep, 2017
  • Bangabandhu Sheikh Mujibur Rahman Digital University, Bangladesh

How to determine sample size in quantitative research?

What is the justification of its validation?

Most recent answer

11th Mar, 2022
Jose Roberto Rasi
FACCAT
The sample size is calculated using a power analysis. A power analysis calculates, for varying sample sizes, a probability (power, β) of finding a statistically significant result (at chosen Type I error, α) for a given population effect size (Cohen, 1988).
1 Recommendation

Popular Answers (1)

2nd Jan, 2018
Haitham Hmoud Alshibly
Al-Balqa' Applied University
Dear RG colleague,
Determining the sample sizes involve resource and statistical issues. Usually, researchers regard 100 participants as the minimum sample size when the population is large. However, In most studies the sample size is determined effectively by two factors: (1) the nature of data analysis proposed and (2) estimated response rate.
For example, if you plan to use a linear regression a sample size of 50+ 8K is required, where K is the number of predictors. Some researchers believes it is desirable to have at least 10 respondents for each item being tested in a factor analysis, Further, up to 300 responses is not unusual for Likert scale development according to other researchers.
Another method of calculating the required sample size is using the Power and Sample size program (www.power-analysis.com).
Regards,
18 Recommendations

All Answers (41)

11th Sep, 2017
Bruce Neagle
PS Biotech
Statistically (ISO 2859?)!  There was a discussion on Linkedin.
11th Sep, 2017
Alan F Rawle
Malvern Panalytical
Depends on how close you wish to be to the 'truth'...
11th Sep, 2017
Julian Nasello
Psychiatric Day Hospital "La Clé"
Hi, 
You have many websites that propose to calculte sample size:
And here it is a paper answering to your second question: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2876926/
I hope that will help you,
Julian
12th Sep, 2017
Alma Braimllari Spaho
University of Tirana
The sample size must be 5 to 10 % of the popullation and in the case of  a small population (less 100) all the population. 
2 Recommendations
12th Sep, 2017
Jan Ivanouw
University of Copenhagen
Which sample size is needed depend on the analyses  you make, and which statistics you focus on. Different statistics from the same research implementatin may require different sample sizes.
If you ar using the Mplus program for you analyses (which are recommended, as this is a very, very flexible and easy program with a lot of support material), you can determine sample size using Monte Carlo techniques.
See: Muthén, L.K. & Muthén, B.O. (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 4, 599-620.
13th Sep, 2017
Kenneth Lui-ming Ngie
The University of Newcastle, Australia
The following sources are also helpful: 
  • Burmeister, E. and Aitken, L. M. (2012). Sample size: How many is enough? Australian Critical Care, 25, 4, pp. 271-274.
  • Lenth, R. V. (2001). Some Practical Guidelines for Effective Sample Size Determination. The American Statistician, 55, 3, pp. 187-193. 
  • Sample Size Calculation? : https://www.researchgate.net/post/Sample_Size_Calculation 
2nd Jan, 2018
Haitham Hmoud Alshibly
Al-Balqa' Applied University
Dear RG colleague,
Determining the sample sizes involve resource and statistical issues. Usually, researchers regard 100 participants as the minimum sample size when the population is large. However, In most studies the sample size is determined effectively by two factors: (1) the nature of data analysis proposed and (2) estimated response rate.
For example, if you plan to use a linear regression a sample size of 50+ 8K is required, where K is the number of predictors. Some researchers believes it is desirable to have at least 10 respondents for each item being tested in a factor analysis, Further, up to 300 responses is not unusual for Likert scale development according to other researchers.
Another method of calculating the required sample size is using the Power and Sample size program (www.power-analysis.com).
Regards,
18 Recommendations
2nd Jan, 2018
Haitham Hmoud Alshibly
Al-Balqa' Applied University
Dear RG colleague,
Determining the sample sizes involve resource and statistical issues. Usually, researchers regard 100 participants as the minimum sample size when the population is large. However, In most studies the sample size is determined effectively by two factors: (1) the nature of data analysis proposed and (2) estimated response rate.
For example, if you plan to use a linear regression a sample size of 50+ 8K is required, where K is the number of predictors. Some researchers believes it is desirable to have at least 10 respondents for each item being tested in a factor analysis, Further, up to 300 responses is not unusual for Likert scale development according to other researchers.
Another method of calculating the required sample size is using the Power and Sample size program (www.power-analysis.com).
Regards,
1 Recommendation
4th Jul, 2018
Abdullah Noori
Kabul University
This formula is the easiest way to do it.
2 Recommendations
23rd Jul, 2018
Igor Gurkov
National Research University Higher School of Economics
Dear Md. Ashraf Uzzaman ,
Go to http://www.raosoft.com/samplesize.html and you get immediately the sample you need.
Bests,
Igor Gurkov
15 Recommendations
21st Aug, 2018
Bendu M Dawolo
Capella University
Awesome information. Another helpful calculation table can be found at https://www.surveysystem.com/sscalc.htm
1 Recommendation
21st Aug, 2018
Jan Ivanouw
University of Copenhagen
The problem with calculations as the above mentioned is that required sample sizes are different for different types of analyses, and for different types of effects.
In serious research you need to specify the analysis and make the required sample size calculations accordingly.
26th May, 2019
Ana Maria-Irina
Bucharest Academy of Economic Studies
Hello everyone. I would have a question if you can please help. What would be the most representative sample or the best method to calculate this sample if I want to analyze the impact of Social Media in Tourism among Generation Y who are let's say approximately 3.5 million people? Thanks!
24th Jun, 2019
Olga Morozova-Gelmici
University of Liverpool
Alma Braimllari Spaho could you please explain why 5 to 10 % is a good sample? I don't seem to be able to find any literature on this regard. Thank you!
24th Jun, 2019
Olga Morozova-Gelmici
University of Liverpool
Bruce Neagle do you have the link to that discussion by chance? Thank you!
24th Jun, 2019
Bruce Neagle
PS Biotech
Unfortunately no.
30th Jul, 2019
Shajib Ullah
University of Bolton
Hi i m following mixed methodology , i will do survey and semi structure interview, whats the sample size should be ?
2 Recommendations
23rd Oct, 2019
Amee Joan
University Malaysia Sarawak
I am doing a quantitative study. The total population is 2320. What should be the adequate sample size?
4 Recommendations
23rd Oct, 2019
Akarsu Bayram
with 95% confidence level and 5% error, it should be around 330
2 Recommendations
23rd Oct, 2019
Alan F Rawle
Malvern Panalytical
Simply, the sample size is related to the accuracy of the result you require. If you require 1% accuracy on the 'true' mean value then you'll need 10000 (random) participants, particles, experiments etc. The standard error is inversely proportional to the square of the number (of whatevers). Or the number of experiments, participants etc is inversely proportion to the variance (or the standard error squared).
Amee Joan For 0.01 (1%) standard error on the mean you'll need to sample them all. For 0.05 (5%) then 1/202 or 400 is a good guideline and in line with Akarsu Bayram above.
2 Recommendations
23rd Oct, 2019
Amee Joan
University Malaysia Sarawak
Alan F Rawle Thank you
Akarsu Bayram Thank you
1 Recommendation
17th Feb, 2020
Roshan Panditharathna
University of Bolton
For PLS-SEM, better to go for 200 samples for statistical significance. However, it does not matter all the time as 10 time rule is applicable for PLS-SEM, that supports for relatively a smaller sample size.
For covariance based SEM does need at least 300 samples for a decent statistical significance.
2 Recommendations
29th Apr, 2020
Mehboob Ul Hassan
Virtual University of Pakistan
Results of regression analysis shows that overall 67% teachers’ self-efficacy have effected on students’ academic achievements, while sub-factor: students’ engagement 58%, instructional strategies 65% and classroom management have affected 60% on students’ achievement scores. Why there is so variations in effect when we sum 58+65+60= 183/3=61 %, why it is not 67%?
2 Recommendations
5th Sep, 2020
Anjul Gagwani
University of Salford
Can we go with 50 respondents for quantitative survey? Will it be valid for statistical result?
3 Recommendations
5th Sep, 2020
H.M Akramin.
Universiti Brunei Darussalam
am in a similar position, I want to do something based on primary teachers but am not in a position to survey all of the primary teachers in the area due to logistics. Am looking at 100-150 for a 24 item quantitative survey. Not enough people?
2 Recommendations
7th Sep, 2020
Adnan Majeed
Beaconhouse National University
3 Recommendations
4th Oct, 2020
Adnan Majeed
Beaconhouse National University
4th Oct, 2020
Md. Ashraf Uzzaman
Bangabandhu Sheikh Mujibur Rahman Digital University, Bangladesh
Adnan Majeed Thank you
1 Recommendation
28th Dec, 2020
Anuradha Iddagoda
University of Sri Jayewardenepura
Sample Size Rule
Sekaran (2013) wrote:
"Roscoe (1975) proposes the following rules of thumb for determining sample size:
1. Sample sizes larger than 30 and less than 500 are appropriate for most research.
2. Where samples are to be broken into sub-samples;(male/females, juniors/seniors, etc.), a minimum sample size of 30 for each category is necessary.
3. In multivariate research (including multiple regression analyses),the sample size should be several times (preferably 10 times or more) as large as the number of variables in the study.
4. For simple experimental research with tight experimental controls (matched pairs, etc.), successful research is possible with samples as small as 10 to 20 in size."
Reference
Sekaran, U., 2003. Research methods for business: A skill building approach. John Wiley & Sons.
13 Recommendations
28th Dec, 2020
Alan F Rawle
Malvern Panalytical
You can only sum variances, not s.d.’s...
2 Recommendations
18th Apr, 2021
Moreblessings Mapolisa
Zhejiang Gongshang University
rule of thumb
18th Apr, 2021
Bruce Neagle
PS Biotech
As I recall the AOAC (Association of Official Analytical Chemists) had a statistical guide for sample size and analysis.
3 Recommendations
2nd May, 2021
Md Jahangir Alam
University of Dhaka
Thank you so much
17th Jun, 2021
Abdulla Aldarayseh
Emirates College for Advanced Education
21st Jul, 2021
Supil Udash
Nepal Open University
2 Recommendations
21st Jul, 2021
Mehboob Ul Hassan
Virtual University of Pakistan
Thanks Supil Udash for answer
29th Sep, 2021
Adamu Abdullahi Garba
Universiti Teknologi Malaysia
21st Nov, 2021
Amankwa Eric
Grand Canyon University
For sample size calculation you may use Yammes formula or a software called G*Power Analysis .
9th Mar, 2022
Jackie Raborar
University of Santo Tomas
Hi can I read full?

Similar questions and discussions

Related Publications

Article
Full-text available
This editorial is on statistical sampling, which is one of the most two important reasons for editorial rejection from our journal Turkish Librarianship. The stages of quantitative research, the stage in which we are sampling, the importance of sampling for a research, deciding on sample size and sampling methods are summarised briefly.
Article
Full-text available
The 2018 Quantitative Research Training Project offered language teachers based in Japan an opportunity to learn about quantitative research methods, get practical experience doing classroom-based research, and collaborate throughout the process with a group of their peers. Four teachers share their insight from their experience doing this research...
Got a technical question?
Get high-quality answers from experts.