126.How to calculate sample size for my study

[@alms59]

Hallo mam! Thanks a lot! I am listening your lectures first before Swayam for understanding.I am 65 yrs Professor of Respi medicine getting great help from your Videos on such a dry theoretical Topics for BCBR exam.

[@DrSujathasENTClasses]

@@alms59 thank you. Good luck Professor

[@kgowtham004]

thank you so much mam for your wonderful class!! the way you explain and teaching in simple understandable way is awesome!! i hope ill do my exams well :)

[@DrSujathasENTClasses]

🙂good lucks

[@Akshaykumar-dm1ui]

Nice demo mam

[@ganeshtevarannavar4191]

Hello mam good morning, actually i am doing pretest and postest control group design,is an experimental study, then which formula best for calculating sample size

[@banglaobangladesh6954]

excellent mam

[@sanvendrasingh6492]

Good explanation, ma’am I am planning to do a cross sectional study and using a stratified sampling technique. I have total population number, but my proportion is very low I.e 0.00042 . Not getting how to calculate the sample size . There are budget and resources constraints too. I can do study on around 60 -70 sample only. Please let me know how to proceed

[@sanvendrasingh6492]

My objective is to assess the readiness of health and wellness centre for quality certification in all states and UT’s in India. Found only one study similar to it conducted in one state only. Not getting sample size I have all the numbers population and out of population how many got certificate. Proportion is is very low.

[@DrSujathasENTClasses]

To assess the readiness in national survey you can use this formula for sample size calculation n={[ ( z2 *p*q) + ME2 ] / [ME2 + z2*p*q/N] }*d n = sample size, z = confidence level at 95% (1.96), ME = margin of error , p =anticipated proportion, q=1-p, N population size, d = design effect

[@mesiazul8435]

Is it valid to use the mean and standard deviation from a similar, but not exactly the same, study to estimate the sample size for a retrospective study? For example, the reference study was done in patients with diabetes, but the retrospective research will be done in patients who had a stroke.

[@DrSujathasENTClasses]

You have to take similar studies in stroke.

[@augustinesdb]

Great

[@DrSujathasENTClasses]

🌹

[@ramalingamkc2153]

Thank u mam

[@subha1450]

Mam, i plan to do an interventional study-one grp pre test-post test design. Which formula shall i use to calculate the sample size for my study? Pls tell me mam

[@DrSujathasENTClasses]

Is the study design doing comparison between two groups on the basis of selected attributes such as knowledge,perception, and attitude? then formula is Sample size( n) 𝑛=(𝑝1(1−𝑝1)+𝑝2(1−𝑝2))/((𝑝1−𝑝2)2 )*C n = Sample size for one group that we need to find out p1and p2= Proportion of two groups C = Standard value for the corresponding level of α and β selected for the study

[@iamvarunverma]

Mam, I have a question I am going to start a prospective study on DM patients. I am confused about the value of (P) which I should take. General the value of P = 0.5 But I checked the prevalence of Diabetes in India is 9.6% ( from Indian Diabetic Report 2021) with 77 million active cases of DM. Mam, please clear my doubts about this and which value should I use to calculate exact sample size for my study.

[@DrSujathasENTClasses]

For sample size calculation, p is what you are going to find out by your study. Don't confuse it with statistical significance p value. Decide your research question. Then do literature search of studies of similar kind. Take a smilar study as reference study for sample size calculation. For eg, if you are going to find out proportion of complications of DM, take your reference study. List out all complications given in that. Take the complication with lowest percentage as your p to give in your sample size formula ( for descriptive studies it is usually 4pq/d square). Hope it is clear

[@iamvarunverma]

@@DrSujathasENTClasses Mam, I checked similar studies but they generally used either P= 0.5 or used this The incidence and prevalence of DM in the Indian population were considered for the calculation of sample size. . Mam my project is Treatment Satisfaction on DM patients. (Prospective Observational study) . I checked the current prevalence of Diabetes in India which is 9.6%. I calculated via this S=z²*p*(1-P)/M² where Z=95% =1.96 P=9.6% = 0.096 Q = 1-P M² = 5% = 0.05 by using this formula my sample size came out 133 But if I use P=0.5 The sample size came out 384 . Mam, please help me in finding the correct sample size it would be a great help for me.

[@DrSujathasENTClasses]

@@iamvarunverma it is the first one,using prevalence of DM

[@shailendramosamkar8986]

Hello Madam my study design will be prospective cohort,longotudinal study ,so for sample size calculation,my seniors experts asked me to use incidence to calculate the sample size and not to use prevalence.so what formula will be there to calculate the sample size.plz help me.

[@DrSujathasENTClasses]

From literature review,get your parent study. Then take prevalence from that as p. Can use formula as 4pq/d square

[@shailendramosamkar8986]

@@DrSujathasENTClasses Thanks Madam, by the help of literature review of parent study I will take prevalence and use the same formula given by you to calculate the sample size ,Thank you so much for your kind help .if I need any help in future, I will contact you. Once again thanks.

[@maqboolmajani4503]

Hello mam, please help me.. I am doing a research on cadaveric body in which I am measuring distance of nerves from anatomical landmarks.. but I am confused that how much my sample size should be . Because in previous studies the sample size is between 50 to 100. But when I am calculating sample size it is 384. But it is not possible to obtain that number of cadavers.

[@maqboolmajani4503]

I can do on 60 cadavers

[@DrSujathasENTClasses]

sample size calculation depends on study design. If this is a descriptive study,take 4pq/d2. p to be taken from previous studies as per your research question.

[@maqboolmajani4503]

@@DrSujathasENTClasses my study is observational

[@jayantpanvelkar758]

Ma'm u could have given an example of a research and explained the concepts...very difficult to understand

[@DrSujathasENTClasses]

yes, research methodology is a bit confusing

[@krishanthichampika7562]

Dear mam My research study regarding knowledge,attitude and practice regarding antenatal care among husbands attending ANC in particular hospital.( Non experimental descriptive study) There were no similar study ifound in my country to determine p value Can I use prepared table values for p and d to set my sample size 250

[@DrSujathasENTClasses]

If in India, there are lots of similar published articles. If among SAARC countries you can use these