Hi Jasmine

Wow thats a big question - obviously depends on what your data is, but firstly before analysising your data you need to consider weighting your data (for potential responent bias) and then you need to use some sort of survey procedure (eg Proc survey in SAS or Complext survey in SPSS or svy in STAT).

I do population health monitoring and thought this article may may help you to start with - it should be available on my researchgate profile.

Article: Methods of the NSW Schools Physical Activity and Nutrition Survey 2010 (SPANS 2010).

L L Hardy, L King, P Espinel, A D Okely, A Bauman

best wished

lou

The statistical analysis depends on the objective of the study. The objetive of a survey is to obtain information about the situation of the population study. The first statistical task therefore is to do a descriptive anslysis of variables. In this anslysis it is necessary to present results obtained for each type of variable. For qualitatrive and dicotomus variables, results muts be presented as frequencies and percentages. For quantitative varibles, the presentation is as means and deviations. After this analysis, you can assess the association between variables, and predictive analysis based on multiple regression models.

I agree with Lousie answer, just using statistic software can be help for analyze large population. For analyzing your data you must consider type data and statistical analysis model.

regards

A good method to know what analyses are relevant is to review papers for your similar topic. The methods from papers are good reference for you. You will have more ideas after looking many papers.

I agree with Jason. How did you arrive at your sample size from the beginning? What are the aims of your study? You will need to review articles related to your study and that will give you the idea of what analyses need to be done and what softwares were used. You can use SPSS, EpInfo, STATA, Minitab, OpenEpi, Graphpad and many others depending on your useage and familiarity with the software.

In addition to the above respondents' suggestions, you should start by looking at the distributions of age, gender, race/ethnicity, and any measures of socio-economic status that you have (income, education level, access to medical care). These distributions will help to inform your analysis, in terms of possible age-adjustment, weighting, and other analytical tools available to address issues of bias and non-representative samples.

To complete this part of the exercise, retrieve the demographic data collected from the most recent national or regional census, with respect to the distributions of age, gender, etc. Compare these to your study's distributions (e.g., via Chi-square), to see if they're significantly different than that of the population from which your study sample was drawn.

Thank you all for your prompt answers. My survey is on Prevalence of menstrual disorders and awareness of menstrual hygiene among young and adolescent girls of hostel and rural areas. Your suggestions will be of great help and I will review other papers related to such work.

Ok do as suggested above, re explore your data. It sounds like you potentially have lots of categorical, rather than continuous data. You need to consider generating some categories such as age groups, rural v hostels, other sociodemographic characterstics eg socioeconomic status,etc and your outcomes - ie categories menstrual disorders and awareness. Once you have your categories then do some cross tabs to explore where there are differences according to say age group and residence (hostel v rural) etc and type of menstrual disorder. The chi square will calculate the p value and odds ratio. If you need to control for various covariates you will need to use logistic regression models. If continuous data first explore via anova, and then again if you need to control for other factors, then use multiple regression.

Hope this helps - good luck!

cheers

Lou

Thanks a lot Louise... will do as directed

I think Louise has very well explained the statistics applied for your data. As you have to find the prevalence and awareness regarding menstrual hygiene, this is purely categorial data except age group and so. First you have to prepare frequency tables for socio-demographic characteristics and other variables collected. For awareness, you need p-value for stating any statistical association between two variables. You can apply regression too.

Jasmin ji, Most of the answer has been given above by Louise,Naresh Makwana and others. In addition these you can apply simply the descriptive statistics for uni variate analysis and inferential statistics for bi variate and multivariate analysis.

Statistical analysis is depending on the type of variable.

As you have survey is most probably cross sectional type of study.

After deciding type of data and variables u can do data analysis with help of statistical software or manually.

Thank you Dr. Pradip. Software will be more convenient as the population to be studied is large.

The suggestions offered are very good. Your question is very broad. In addition, you need to frame your research questions according to the aims of your study. Your instrument should provide the necessary data to answer your research problems. A sample of your questionnaire or instrument together with your research problems would be of great help. This will allow statisticians or people who has background on statistics to recommend the appropriate statistical method. It would be difficult to suggest a specific statistical tool based on presumptions on the direction of the study that you are undertaking.

May you also think of teaming up with a statistician-scholar, for a better analytical interpretation of your (primary, raw) data. That surely will make significant value addition to your research paper(s). Possible/ appropriate methodologies are of course suggested already, at length.

All the very best.

Input from the statistician will be of immense help in the conduct of the study and will also help in publication of the study in the reputed journal, nowadays there are various ready made software available in the  market . SPSS is widely used in the statistical analysis and accepted by many reputed journals.

Hello All, I have the similar question. What statistical analysis should I use if want to look at the difference in length of stay for a specific surgery among groups of patient (ie. race groups, insurance type, different age group etc.)

thank you for all

You can use different software but SPSS seems to be more common in this case and apply different statistical tools based on your research questions and objectives.Otherwise, you can also make a reference to literature's on similar studies or subjects

Dear Jasmine Ritesh Gujarathi ,

Hi,

You may study the populations genetically using marker; in this case you can use Popgene software for population genetics or you can use STRUCTURE for population genetic structure, but if you have only studied population diversity, you can use PAST, NTSYS-pc, CANOCO & CANO-Draw.

Best,

Saeed

If you are open source user, you can try BiodiversityR package in R program

Best,

Saeed