-
1. Statistical Objectives: State the statistical objectives of the paper in context. For example, it might be a study to compare the effects of two different drugs on cholesterol levels. Another example might be to compare the effects of cell phone use on driving distraction measures.
2. Population: Define the parent population from which the individuals were collected. Is it a sample taken from all students in a single school, from all college students, from all people in a particular country, from all trees from a particular forest? The population defines the group of all possible “individuals” in a sample. Who are the individuals and what assumptions are being made about the individuals in the population? Do you think the assumptions are reasonable? Why or why not?
3. Sample: Statistics are computed on the basis of one or more samples. Describe the set
of all possible samples and the process that was used to collect the sample from which
the data was obtained. What is the size of the sample? Was the sampling process
reasonable in that the assumptions of independence were met, was it random, etc.?
4. Variables and Data Collection Techniques: What variables describe the sample
data? Identify the types of the variables (categorical or continuous). If multiple
variables are being investigated, describe them. Identify which variables are
dependent and independent. For example, if we are studying tree growth, rainfall may
be an independent variable and “growth” would be the response variable.
5. Graphical Displays and Tables: Identify the graphing techniques used (histograms,
normal probability plots, scatter plots for two variables, etc.). Were the visual displays
effective in conveying the desired information or might other graphical techniques or
displays have been more effective? Describe these tables/graphs and how they relate to the objectives.
6. Study Type and Analyses: Identify any associated statistical analyses used that we
have covered in class. For example, is your article describing the results of a designed
experiment or a survey? Describe the nature of the statistical study and identify any
randomization techniques used. Describe any analyses used to construct
p-values or confidence intervals. Are sources of bias present? Might other types of
analyses have been better, and if so, why?
7. Interpretation: In your own words explain the interpretation of the statistical analysis
to someone who knows nothing about statistics. Please don’t cut and paste sentences from the original article. Points may be deducted if you do.
Thursday, March 7, 2013
Statistical Review
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment