Comparing Proportions Overtime, Self and Vis-a-Vis World

#1
Dear Friends,
I am trying to assess the research performance of a Country between 2010 and 2016, through the quality of the research papers of the country. The TOTAL published papers for the Country and the World were classified into 4 categories (Q1: high Quality, Q2: second tier quality, Q3: 3rd Tier and bottom tier quality: Q4) as indicated in the attached table.

I would like to test:
1. The performance of the Country between for the high quality research [i.e., proportion of Q12010 (28.7%) and proportion of Q12016 (33.2%)] and the low quality research [i.e., proportion of Q42010 (17.4%) and proportion of Q42016 (15.9%)].

2. The performance of the Country vis-à-vis the world (between proportion of CountryQ12010/WorldQ12010 (1.59%) and CountryQ12016/WorldQ12016 (2.71%)), and the same for Q4 (1.79% and 2.97%).

Any suggestions on which test is applicable in this case? Please note that the numbers are totals and not a samples.

All the best Taisir
 

Karabiner

TS Contributor
#2
I would like to test:
1. The performance of the Country between for the high quality research (...) and the low quality research
What do you mean by this? Wheter the respective proportion changed between 2010 and 2016? Whether the proportion (pooled for year 2010 to 2016) of high quality research is larger than the proportion of low quality research?

With kind regards

Karabiner
 
#3
What do you mean by this? Wheter the respective proportion changed between 2010 and 2016? Whether the proportion (pooled for year 2010 to 2016) of high quality research is larger than the proportion of low quality research?

With kind regards

Karabiner
Dear Karabiner
Respective proportions between 2010 and 2016 for Q1 and Q4, not pooled.

I am trying to compare the high quality proportion of 2010 with the high quality proportion of 2016; and
low the quality proportion of 2010 with the low quality proportion of 2016
as mentioned in 1 of the post

as for the country versus world,

I would like to
compare the contribution/proportion of the country in high quality between 2010 and 2016, and
compare the contribution/proportion of the country in low quality between 2010 and 2016,

I hope that helps

as mentioned in 2 of the post


All the best and many thanks in advanced Taisir
 

Karabiner

TS Contributor
#4
I am trying to compare the high quality proportion of 2010 with the high quality proportion of 2016; and
low the quality proportion of 2010 with the low quality proportion of 2016
as mentioned in 1 of the post
Well...no, you did not, actually.- Anyway, if you have data from years 2010 and 2016, and data tell you how many (i.e. actual frequencies, not proportions) papers were of high quality and of low quality, respectively, you can construct a 2x2 table with "year" and "quality" and perform a Chi² test.

Regarding the contribution to the world's output, you can construct a 2x2 table for high quality publications, with "year" (2010 vs. 2016) and "origin of the paper" (specific country vs. rest of the world). Again, you would use frequencies, not proportions.

With kind regards

Karabiner
 
#5
Dear Karabiner,
Many thanks for your prompt response and advice. Indeed, I have original frequencies and will try your suggestions with Chi Square test.
My apologies for not communicating the problem clearly.

Will keep the thread open for a couple of days for others to contribute if they wish.

Taisir
 
#6
Dear Karabainer
Greetings again.

Per your suggestion, I constructed the 2X2 tables and performed Chi Square calculation, to the best of my knowledge.

I would appreciate it if you can look at the attached and let me know if I am on the right track or not.

Once confirmed, I plan to repeat the process for a number of countries.

Many thanks and I appreciate your help. Taisir
 
#7
Dear Karabainer
Greetings again.

Per your suggestion, I constructed the 2X2 tables and performed Chi Square calculation, to the best of my knowledge.

I would appreciate it if you can look at the attached and let me know if I am on the right track or not.

Once confirmed, I plan to repeat the process for a number of countries.

Many thanks and I appreciate your help. Taisir
Dear All
I Would appreciate other members help on this thread.
Many thanks Taisir