Internal Code: 1GHBB
Statistics Report Writing Assignment Help:
Task:
Data collection
Collect quantitative sample data on two variables. Ideally you will be observing two characteristics for each individual in your sample. It is not a strict requirement but if you have bivariate data you will be able to use this same data for assignment 2 later in the course.
The data collection is not the main focus of the assignment and therefore you should not spend a great deal of time on this. You can either generate the data yourself via survey or use data that is already published. The data collection method does not need to be perfect. Indeed, weaknesses in the data is one aspect you should report on.
Ultimately you want two sets of numerical data that you can calculate sample statistics for - particularly, either the sample mean or sample proportion for each set. To ensure valid inferences can be made, you should have at least 30 observations for each variable.
Data description
Using the methods we have studied, describe your two data sets and any relationship between them. Your discussion should include
1. what type of variables you have obtained data for
2. how you obtained the data / where the data came from
3. appropriate graphical representations of the data
4. appropriate numerical summary measures for the data
Highlight the important features and comment on aspects you found surprising or interesting.
Data analysis
Consider the two populations you have sampled from. Identify one key parameter of interest for each population - this will either be the population mean or the population proportion. Without taking into account any information from your sample data, pick a value for each parameter that you think is reasonable. Assuming those parameter
values are correct, derive the sampling distributions for the estimators of those parameters. I.e. if you have two assumed population proportions, derive the sampling distributions for the sample proportions, or if you have assumed population means derive the sampling distributions of the sample means. In the latter case, you can assume that the population standard deviations are equal to the sample standard deviations you obtained. Now, answer the following question:
a. How does the distribution of the underlying population compare with the sampling distribution for each variable?
b. How likely is it that you would observe the sample statistics (means or proportions) you have obtained (or something more extreme) if your parameter assumptions were correct? (I'm just asking you to find a p-value for each sample statistic here)
Use your sample statistics, construct confidence intervals for your parameter of interest from each data set. Comment on any implications of your estimated confidence intervals. E.g. do they overlap? do they contain the parameter value you assumed to derive the sampling distribution? do they contain zero?
Assignment output
Submit a report encompassing the above as a MS Word document and attach any related working in a single MS Excel workbook. A suggested target range for the word count of the document is 1000 -1500 words.