Thread Closed 
Week 2: Nations' Ranks from 1960 to 2004 in Infant Mortality Rates
09-07-2014, 09:02 PM
Post: #11
RE: Week 2: Nations' Ranks from 1960 to 2004 in Infant Mortality Rates
What seems right: Countries are ranked from lowest infant mortality rate at the top, and highest at the bottom. It is easy to see the relative rankings of the countries listed, and whether their relative ranking went up or down between the two time periods. I personally like the lines so it is easy to locate the ranking pair for a single country between the years. The United States is easy to locate with the bold font and red line. I like how the line color is different depending on whether the relative ranking increased or decreased.

What seems wrong: The line crossings are dense. More color should be used to make it easier to distinguish the lines. Perhaps solid color could be used for increasing and dashed lines for decreasing. The rankings are only relative. We don't know if the real death rates increased or decreased. It could be that all the rates increased dramatically for all countries. It is also possible that for a particular country the rate stayed constant, but its ranking position changed because of the change of rate in other countries. We don't know because we are not given actual numbers. There are many countries missing, but there is no explanation of why these particular countries were chosen while excluding many more countries. We are not given the criteria for which countries are represented. And on that note, we are not told what the definition for infant mortality rate is. Does the rate include deaths starting at conception, or does the infant need to survive to a particular age first (such as 1 month of age), and what is the cutoff age? Did this criteria stay the same between the two time periods? Issues such as these have arisen in other medically related comparisons. It is important to be precise.

I agree that interactivity would improve this visualization. Highlighting a country when moving a mouse over it would be good, and then extra information could be displayed. I agree that the lower/higher labels are distracting. When I first read the "Lower" label I was confused since I thought it meant the line with the negative slope meant the rate had decreased from 1960 to 2004, but in fact the relative ranking had increased.
Find all posts by this user
09-07-2014, 09:59 PM
Post: #12
RE: Week 2: Nations' Ranks from 1960 to 2004 in Infant Mortality Rates
lediaev makes an excellent point. The method for determining the infant mortality rate needs to be defined. For all we know, each country might make the measurement differently or the definition may have changed between the two measurements.
Find all posts by this user
09-08-2014, 10:15 PM
Post: #13
RE: Week 2: Nations' Ranks from 1960 to 2004 in Infant Mortality Rates
I think the clearest information this chart gives us is that the rank of United States declined obviously in 2004. This design could attract readers' focus effectively and impress people. However, there are still many design problems. Like previous discussion, scale is not mentioned. The rank of countries and specific infant death rate of these countries are totally two different conceptions. The decline of the rank can't represent that the rate of infant death increases. Similarly, Singapore, HongKong and Japan rise to top 3 can not explain that infants get higher survival rate. So one of the important elements of visual design -- the baseline of the data is not supplied.
Also, the contrast of each country is so vague. Light and dark lines twine together at the center of the chart that we have to apply great efforts to recognize. Maybe a good method to improve it is that each country use two
bars to represent its different years' infant death rates, so it will be very clear whether the rate declines or rises, then give each bar the rank number which could indicate the country's condition in the whole world.
Find all posts by this user
09-09-2014, 11:13 AM
Post: #14
RE: Week 2: Nations' Ranks from 1960 to 2004 in Infant Mortality Rates
As my opinion, ths designer of this graph might want this graph sample enough, and i can clearly see how the rank of infant death rates for each country changed from 1960 to 2004, but there are some points which i think is not good enough, first, from this graph, i cah only know how the rank changed, but have no idea about the number for each country rank, i dont wanna count for each country. And second, the color of the line are similiar, sometimes, i can not figure out what the line point at. But if i just want to know how the rank changed, such as, it rank higher or lower, so, it is a clearly graph.
Find all posts by this user
09-09-2014, 01:43 PM
Post: #15
RE: Week 2: Nations' Ranks from 1960 to 2004 in Infant Mortality Rates
I agree with Kevin that a great thing about this visualization is its high data to ink ration. The biggest problem I found is scale distortion. Just by looking at the ranking we can't figure out whether infant mortality had improved or not. Since rankings are relative, US mortality might have dropped while its ranking went down. Also, when I first saw it I was confused as to why they put lower/higher infant mortality rate in the middle of the lines. This practice is unusual, so having a vertical bar with numeric value would have clarified this visualization.
Find all posts by this user
Thread Closed 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)