TASK DESCRIPTION In this portfolio item, you are asked to critique a text that contains an example of bad statistical reasoning WORD LIMIT 300-350 Instructions Find a short text/article that contains an example of poor statistical reasoning and provide an analysis (in your own words) of why it illustrates bad statistical reasoning. Your analysis should consist of 250-500 words and discuss one or more of the fallacies or common errors that we have discussed in weeks 9-12. For example, it could illustrate Simpson’s paradox, the Prosecutor’s fallacy, inappropriate assumption of independence, generalising inappropriately from a biased sample, assuming correlation implies causation, regression to the mean. It is okay to reuse an article from the weekly discussion forums. However, your analysis should be in your own original work. USE THE ERRORS WHICH ARE GIVEN IN TASK INSTRUCTIONS……………FOR PORTFOLIO ANALYSIS
EXAMPLE: Analysis of text/article: Statistical reasoning type(s) relevant to this article include “Self-selected samples”, “Confounding variables”, “Cause and effect”, “Randomised experiment”, “Averages”. The implicit argument being made here is that students should attend PASS sessions as it will help them earn higher grades and reduce the risk of withdrawing from a unit. A potential problem with this line of reasoning is that the students who choose to attend PASS sessions are self-selecting and that they might be likely to possess attributes that naturally make them more likely to do well in units. For example, possible confounding variables that might be associated both with attending PASS and earning high grades are having sufficient spare time to devote to study, and intrinsic levels of motivation. As a general principle it is hard to determine cause and effect on the basis of an association unless data come from a randomised experiment or if possible confounding variables can be identified and controlled for. One way to try and control for confounding variables in this situation would be to compare the average grades of students who attend PASS sessions to the average grades of students who do not attend PASS sessions but for units in which PASS is not available. If the difference in average grades remained it would suggest that the observed differences may be due to confounding variables, whereas if the difference disappeared or was less prominent that would suggest that PASS does indeed help improve grades. It does not seem like it would be practically possible to do a randomised controlled trial in this situation!Subject Good Thinking
|
Analysis of text/article: Statistical reasoning type(s) relevant to this article include “Self-selected samples”, “Confounding variables”, “Cause and effect”, “Randomised experiment”, “Averages”. The implicit argument being made here is that students should attend PASS sessions as it will help them earn higher grades and reduce the risk of withdrawing from a unit. A potential problem with this line of reasoning is that the students who choose to attend PASS sessions are self-selecting and that they might be likely to possess attributes that naturally make them more likely to do well in units. For example, possible confounding variables that might be associated both with attending PASS and earning high grades are having sufficient spare time to devote to study, and intrinsic levels of motivation. As a general principle it is hard to determine cause and effect on the basis of an association unless data come from a randomised experiment or if possible confounding variables can be identified and controlled for. One way to try and control for confounding variables in this situation would be to compare the average grades of students who attend PASS sessions to the average grades of students who do not attend PASS sessions but for units in which PASS is not available. If the difference in average grades remained it would suggest that the observed differences may be due to confounding variables, whereas if the difference disappeared or was less prominent that would suggest that PASS does indeed help improve grades. It does not seem like it would be practically possible to do a randomised controlled trial in this situation!Subject Good Thinking TASK DESCRIPTION: In this portfolio item, you are asked to critique a text that contains an example of bad statistical reasoning |
The post TASK DESCRIPTION In this portfolio item appeared first on My Assignment Online.
