There are actionable ways to avoid the use of biased data
written in one line or less
Accessible?
The first section is easy to understand and motivates the problem well. However, the whole example
(starting with "A few years ago") is difficult to understand and not well suited for illustration. I recommend to add more explanation, i.e., what is meant by the link between defect and commit. What is the value of this link? What is the value of the figure? I assume that all this covers the topic "When do changes induce fixes" but this is not explicitly mentioned in the text.
The guidelines are not suited for someone not familiar with statistics but I think that they have an appropriate level
and can motivate the reader to further explore the field.
Size?
Is the chapter the right length?
ok
Should anything missing be added?
I recommend to explain earlier the meaning of feature.
Can anything superfluous be removed (e.g. by deleting some section that does not work so well or by using less jargon, less formulae, lees diagrams, less references).?
I do not see the value of the figure. Maybe you add a concrete visualization to illustrate the first paragraph of the Section "Identifying bias".
What are the aspects of the chapter that authors SHOULD change?
The example (Bird 2009 study) needs better explanation or could be removed by another one.
Gotta Mantra?
We encouraged (but did not require) the chapter title to be a mantra or something cute/catchy, i.e., some slogan reflecting best practice for data science for SE? If you have suggestion for a better title, please put them here.
I think the title fits very well-
Best Points
What are the best points of the chapter that the authors should NOT change?
The guidelines and the example in the first section (i.e., the voting example).
Title of chapter
Don't embarrass yourself: Beware of bias in your data
URL to the chapter
https://github.com/ds4se/chapters/blob/master/cabird/bias-chapter.md
Message?
There are actionable ways to avoid the use of biased data
written in one line or less
Accessible?
The first section is easy to understand and motivates the problem well. However, the whole example (starting with "A few years ago") is difficult to understand and not well suited for illustration. I recommend to add more explanation, i.e., what is meant by the link between defect and commit. What is the value of this link? What is the value of the figure? I assume that all this covers the topic "When do changes induce fixes" but this is not explicitly mentioned in the text. The guidelines are not suited for someone not familiar with statistics but I think that they have an appropriate level and can motivate the reader to further explore the field.
Size?
Is the chapter the right length? ok
Should anything missing be added? I recommend to explain earlier the meaning of feature.
Can anything superfluous be removed (e.g. by deleting some section that does not work so well or by using less jargon, less formulae, lees diagrams, less references).? I do not see the value of the figure. Maybe you add a concrete visualization to illustrate the first paragraph of the Section "Identifying bias". What are the aspects of the chapter that authors SHOULD change?
The example (Bird 2009 study) needs better explanation or could be removed by another one.
Gotta Mantra?
We encouraged (but did not require) the chapter title to be a mantra or something cute/catchy, i.e., some slogan reflecting best practice for data science for SE? If you have suggestion for a better title, please put them here.
I think the title fits very well-
Best Points
What are the best points of the chapter that the authors should NOT change?
The guidelines and the example in the first section (i.e., the voting example).