Introduction to Critical Thinking for Anyone

This is not the right answer. Although true, it does not make Elsa’s conclusion correct. There is indeed a clear difference between the two vegetable baskets, but our data collection suffers from a selection bias. Put simply, people (probably) just took the nice cucumbers and left the ugly ones. More in the abstract, vegetables in a mostly sold-out box are not statistically representative of the vegetables when the box was still full. In the ICTA course, we’ll train recognizing selection bias.

This is not the right answer. Although our personal preferences for ‘perfect-looking’ vegetables can differ, there is a clear, objective difference between the two boxes (left: ripe and colorful, right: rotten and brown), with most people willing to pay less money for those rotten vegetables. The deeper problem is that our data collection suffers from a selection bias. Put simply, people (probably) just took the nice cucumbers and left the ugly ones. More in the abstract, vegetables in a mostly sold-out box are not statistically representative of the vegetables when the box was still full. In the ICTA course, we’ll train recognizing selection bias.

This is the right answer, good job! Our data collection suffers from a selection bias. Put simply, people (probably) just took the nice cucumbers and left the ugly ones. More in the abstract, vegetables in a mostly empty box are not statistically representative of the vegetables when the box was still full. In the ICTA course, we’ll train recognizing selection bias.

This is not the right answer. Although true, it does not make Elsa’s conclusion correct. Elsa probably does have enough data from her experience (hundreds/thousands of buildings) to say that there is a clear difference. However, her data collection suffers from a selection bias. Put simply, people (probably) maintained the pretty buildings more so than non-pretty buildings, leaving us with mostly the pretty buildings from the olden days. With new buildings, we simply see everything that is built. Elsa’s observations cannot directly make out this difference. More in the abstract, buildings that have been maintained and not demolished are not statistically representative of the full set of buildings when they were built. In the ICTA course, we’ll train recognizing selection bias.

Although maybe true, this is not the right answer. Although our personal preferences for pretty buildings can differ, we could imagine everyone sharing her opinion on the matter. The deeper problem is that our data collection suffers from a selection bias. Put simply, people (probably) maintained the pretty buildings, leaving us with only the pretty buildings from the olden days. With new buildings, we simply see everything that is built. More in the abstract, buildings that have been maintained and not demolished are not statistically representative of the full set of buildings when they were built. In the ICTA course, we’ll train recognizing selection bias.

This is the right answer, great job! Our data collection suffers from a selection bias. Put simply, people (probably) maintained the pretty buildings, leaving us with only the pretty buildings from the olden days. With new buildings, we simply see everything that is built. More in the abstract, buildings that have been maintained and not demolished are not statistically representative of the full set of buildings when they were built. In the ICTA course, we’ll train recognizing selection bias.