- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Posted 06-23-2009 06:02 PM
(1858 views)
Hi,
Where can I find some sample statistical reports that people have analyzed before? I have some data on age, year, gender, etc. I need to conduct some statistical analysis. I am not sure where to start.
Thanks
Where can I find some sample statistical reports that people have analyzed before? I have some data on age, year, gender, etc. I need to conduct some statistical analysis. I am not sure where to start.
Thanks
5 REPLIES 5
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hi cosmid,
I suggest you begin by calculating descriptive statistics for your variables. For continuous variables such as age, year etc this will be mean, standard deviation, median, minimum, maximum, first and third quartiles. Plot the histograms to see the distributions. For categorical variables such as gender etc calculate the frequency and percentages.
Then, proceed with examining univariate associations between pairs of variables. Look at the scatter plot and correlation coefficient (Pearson or Spearman, whichever is appropriate) for pairs of continuous variables such as age and year. For continuous and categorical variable pairs use a t-test/ANOVA/non-parametric analysis of variance depending on the nature of your variables (this is why previous step is important).
Further, find out why these data where collected in the first place - what was the purpose of the study? What questions were asked/hypotheses made? Accordingly, proceed with examining multivariate associations among your variables.
Regards,
Olga
I suggest you begin by calculating descriptive statistics for your variables. For continuous variables such as age, year etc this will be mean, standard deviation, median, minimum, maximum, first and third quartiles. Plot the histograms to see the distributions. For categorical variables such as gender etc calculate the frequency and percentages.
Then, proceed with examining univariate associations between pairs of variables. Look at the scatter plot and correlation coefficient (Pearson or Spearman, whichever is appropriate) for pairs of continuous variables such as age and year. For continuous and categorical variable pairs use a t-test/ANOVA/non-parametric analysis of variance depending on the nature of your variables (this is why previous step is important).
Further, find out why these data where collected in the first place - what was the purpose of the study? What questions were asked/hypotheses made? Accordingly, proceed with examining multivariate associations among your variables.
Regards,
Olga
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Thank you for your detailed response. This will be my first statistical analysis using real data in a real situation. I am kind of nervous and I want to make sure I am doing everything correctly. Because I am the only one working on it there is no one that I can really ask for assistance.
These data were collected over 10 years. There was no hypothesis or any questions. The data is consisted on some common variables such as age, gender, year of event, etc. Do you think if it is possible to show you some of my work that I did on the data? Just to make sure I am going in the right direction?
thanks!
These data were collected over 10 years. There was no hypothesis or any questions. The data is consisted on some common variables such as age, gender, year of event, etc. Do you think if it is possible to show you some of my work that I did on the data? Just to make sure I am going in the right direction?
thanks!
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hi cosmid,
You wrote "These data were collected over 10 years. There was no hypothesis or any questions." Are there questions/hypothesis now? What is the purpose of your analysis?
I don't think it's appropriate for me to look at your work and give you anything beyond general advice. I also think it's not appropriate to discuss specifics of your analysis here at the SAS Statistical Procedures forum if the discussion isn't about usage of SAS to do what you want with your data.
Best regards,
Olga
You wrote "These data were collected over 10 years. There was no hypothesis or any questions." Are there questions/hypothesis now? What is the purpose of your analysis?
I don't think it's appropriate for me to look at your work and give you anything beyond general advice. I also think it's not appropriate to discuss specifics of your analysis here at the SAS Statistical Procedures forum if the discussion isn't about usage of SAS to do what you want with your data.
Best regards,
Olga
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
I agree with you. But where can I find help on generating statistical analysis? How do I know if I am doing everything right?
Basically I was given a set of data that include 10 years of data. The data, like i said earlier, consist of just location, gender, age, year of event, etc. Just some really basic information. My objective was to generate some meaningful statistical analysis. So far I only did some Proc Means and Proc Freq. I mean, what kind of more complex analysis can you get from age, gender and location?
I am sorry I am not even asking the right or appropriate questions. I am just so lost here.
Thanks again for all your input and help. I really appreciate it!
Basically I was given a set of data that include 10 years of data. The data, like i said earlier, consist of just location, gender, age, year of event, etc. Just some really basic information. My objective was to generate some meaningful statistical analysis. So far I only did some Proc Means and Proc Freq. I mean, what kind of more complex analysis can you get from age, gender and location?
I am sorry I am not even asking the right or appropriate questions. I am just so lost here.
Thanks again for all your input and help. I really appreciate it!
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
ok, you used proc means and proc freq to calculate descriptive statistics. Now, explore proc univariate or proc capability to plot the histograms.