[Figures] Add captions

[a2975667]

have detailed captions please! It is essential for the reader to know why they are seeing a figure; as I said in the #voting channel in karrie’s slack group, without helping the reader, you are doing yourself a disservice,

[a2975667]

as a quick suggestion, always have detailed captions: the caption should explain: what this is figure about? to what should the reader pay attention? what is the main point / insight? If adding these explanations seem superfluous, then perhaps we should re-consider the figure. Figures should complement the text, and explain / show / clarify ideas in a manner that facilitates insight (i.e. not readily obvious from reading the text).

[tiffanylwt]

Fix the caption issue with the landscape figure.

[a2975667]

<@UJTHB0WJF>: I had a quick look—thank you for updating the captions. We need much more detail. Figures are part of the argumentation in the paper, not something to be skimmed over. For example,

What is interesting about figure 4? to what should reader pay attention to? why bother showing this distribution of absolute amounts? Could we not just report summary statistics?

What makes a figure interesting is that it contributes to the author's argument. You've chosen to show me donation amounts in a very specific way. It does add to your argument (that Likert may bias towards extremes) I would write it like this:

"The figure shows absolute donation amounts for four conditions: Likert (n=?), QV-36 (n=?), QV-108 (n=?), QV-324 (n=?), in a between subjects study. The suffixes refer to voice credits each of the QV conditions. Notice two issues in the Likert condition donation amounts. First, the average total donation amounts for Likert ($ x) are less than QV ($z1, $z2, $z3). Second, LIkert is mostly absent in the middle range $17-28, while donations in the QV condition are well represented across the donation range. "

notice that I haven't suggested using statistical tests (e..g t-test) to imply that these differences are significant. Some papers will do this in the caption, but I would like to avoid standard tests in this paper to avoid confusion. Thus this caption just suggests a trend, that will be amplified later with the bayesian analysis.

You need write something as detailed for every figure. Guide the reader—give them cues that you've thought deeply about your research; make them feel that they've gained insight. All these little things are more obvious for you as a researcher than for the reader.

For figure 5, somewhere in in your caption, point them to the appendix for a detailed description (e.g. screen shot) of the what the subject read as a description for each issue. It's clear that there are differences in donation amounts across issues; as a reader, I would be curious why this is happening. _{View in Slack}

[a2975667]

Closing this for now. It seems like we had some details now. Can wee what we want to emp later on.