« Popularity contests and charts | Main | A luxury we can afford to miss »

Comments

sparklydatepalm

Thanks for the explanation, I wasn't sure why you had a long weekend.

Tom West

More anti-Canadian bias - we just had our long weekend (Victoria day)! Do you wish us all to get skin cancer and die, or do you think we live in a land of snow and ice?
;-)

Brotherk_13

The only slightly "weird" thing about the line chart is the descending lines - maybe I've just been conditioned to see this as an improving trend. For some reason I want to see both the low weeks and the high Vit A occupying the upper quadrant. However, reverse scaling the time makes no sense, and would only confuse a reader. Can't think of a solve for this, it was just my immediate reaction when I was trying to see what you were charting.

floormaster squeeze

I love your work in general but I definitely disagree with your overuse of lines. Lines emphasize the continuity between the units.

It is appropriate to criticize the use of columns for time series data (artificially making something continuous into discrete "time pockets") but I think there is some legitimacy for using discrete categories here. You could argue that there is some continuity but I would need to know more about the subject matter (I am admittedly not expert) to know if you can claim the "control" is the "0% Vitamin A" value. Even if it was continuous your scale then is wrong as the 0.5% value should be over 5 units from the origin (0%).

Also, the original chart correctly allows you make the natural comparison on the weeks scale. Yours, put not including the origin does not (I am sure you have heard this criticism before). Your chart makes it look like there is a lot more change than their is--this is a classic propaganda technique so I am surprised you are not more sensitive to it.

Again, I love your work and your analytic style but you are the one who created more problems than you (potentially) solved here.

Taf

I agree with floormaster about the use of lines. Commonly in drug tests the body's reaction to drugs isn't always linear, and testing with A and A+1 (and even with A+2) shouldn't be used to conclude A + 0.5 is halfway through the impact...

Kaiser

Tom: ha ha

Brotherk: Good point. The idea of reversing the vertical axis did cross my mind briefly but I decided against it. Instead of artificially reversing the axis, I'd have preferred to find a metric that would have the property that larger is better. I'd hate to trade one unintuitive thing with another unintuitive thing (i.e. to have numbers declining moving up the vertical axis).

Floormaster: If you have been following the blog for a while, you'll observe that the bar chart brigade is not very far away. Every time I post a profile chart, there are a few comments saying it's an absolute no-no. One day now, I will convince some of you. I welcome the disagreement nonetheless.

floormaster squeeze

This is not a matter of being in a brigade or not. It is a matter of clarity and accuracy. I may prefer bars or columns for a whole bunch of nonthematic/"irrational" (in truth, it is largely a preference born out of experience) reasons but that is not what my comment was about.

I apologize if I have not heard your argument for adding the continuity of a line (when there may not be any) but even if you use a line you have problems with scaling on both they horizontal axis (0.1% to 0.5%) and vertical axis (not including the origin exaggerates the change in weeks).

Nick Cox

Plotting with depth below surface on the vertical axis -- so that numbers increase downwards -- is utterly standard in soil science, sedimentology, limnology (study of lakes) and oceanography. It's the way scientists see the data in every sense. And sure, everyone knows that it is not the response or outcome variable.

Dev

Skin cancer is becoming one of the biggest killers over here in the UK. surprising considering the weather! Luckily, more and more of our sunblocks appear to have Vit A added..

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Marketing analytics and data visualization expert. Author and Speaker. Currently at Vimeo and NYU. See my full bio.

Book Blog



Link to junkcharts

Graphics design by Amanda Lee

The Read



Good Books

Keep in Touch

follow me on Twitter

Residues