Approximating percentage of academic traffic in the World Wide Web and rankings of countries based on academic traffic
Bahram Kalhor, Mohammad Reza Ghane, Alireza Nikravanshalmani
AApproximating percentage of academic traffic in the World Wide Web and rankings of countries based on academic traffic
Bahram Kalhor , Mohammad Reza Ghane , Alireza Nikravanshalmani1 Department of Computer, College of Mechatronic, Karaj Branch, Islamic Azad University, Alborz, IRANDepartment of Computer Science, Tehran Science Regional Information Center for Science and Technology (RICeST), Shiraz, IRAN
[email protected]; [email protected]; [email protected]
Abstract
The paper introduces a novel mechanism for approximating traffic of the academic sites (universities and research institutes) in the World Wide Web based on Alexa’s rankings. Firstly we introduce and discuss new method for calculating score (weight) of each site based on its Alexa’s rank. Secondly we calculate percentage of academic traffic in the World Wide Web. Thirdly we introduce and discuss two new rankings of countries based on academic traffic. Finally we discuss about three indicators and effects of them in traffic of the academic sites. Results indicate that the methodology can be useful for approximating traffic of the academic sites and producing rankings of countries in practice.
Keywords:
Weight of traffic,
Percentage of academic traffic , Informetrics, Rankings of countries, Traffic of site, Performance Introduction Data and methods
For approximating traffics of countries and estimating percentage of academic traffic in the world, rankings of traffics for more than 21000 universities and research centers have been used in this study.
Selecting countries and universities
In this study our database contains 21,485 universities, which is the same as the universities announced in to web rankings of world universities (WR, http://webometrics.info). WR is a web based rankings of universities which contains broad list of universities from all over the world (January 2014, 21,451 universities). All countries which have at least one university in our database are considered in this study. 21,485 sites of universities and research centers have been covered by 197 countries and 1 international category (5 universities). hese numbers seem too large in some cases. For instance, Finland has 16 universities and 29 professional university colleges
Weight of universities
Alexa only provides the rankings of sites. It doesn’t give the computed score of each site that has been used for their ranking method. Weighting universities is utilized to approximate the number of hits. It is also used to rank of countries and to compute academic traffic to this end. W u is proposed as the new indicator which is used to detect the weight of each university. W u =1- R u M where W u is weight of university, R u is global rank of university in Alexa and M is maximum rank of sites according to Alexa (M=30,000,000), Min W u = W u =1-(1/30,000,000) =0.9999999666667. Weight of countries
The proposed new indicator W u is then used to calculate the weight of each country. The new formula for this purpose is proposed as follow: W c = ∑ W u i ni=1 where W c is country’s weight, n is the number of universities in the country and W u i is the weight of i th university. Weight of academic traffic
Academic traffic is computed according to weight of all universities. W at = ∑ W u i ki=1 Where W a is weight of academic traffic, k is number of universities in the world (21,485 in this study, January 2014), W u i is weight of university number (i). 𝑃 𝑎𝑡 : Percentage of academic traffic Finally, we introduce P at as a new indicator for calculating academic traffic in the world, based on universities rank in Alexa and maximum rank of 30 million sites in Alexa. The proposed formula is as follows: 𝑎𝑡 = W at ∑ *100 where M is the total number of sites according to Alexa, W at is the total weights of universities, ∑ is the total weights of sites and P 𝑎𝑡 is the percentage of academic traffic. 2.7. Average weight of countries
New formula has been proposed for comparing countries based on number of universities and academic’s weight of countries. Average rank of each country has been calculated by dividing the weight of country to number of universities. Table (4) shows the result of average weight of countries (average traffic rank) for all countries which have at least 100 universities. A wc = ∑ 𝑊 𝑢 𝑖 𝑛i=1 𝑛 where n is the total number of university in each country , W 𝑢𝑖 is the weight of university, A wc is the average weight of country. Results
Table (1) shows the top 60 countries with highest number of active universities. There are some differences between real universities count in each country with data of table (1) which has been collected from webometrics.info. For instance, Finland has 16 universities and 29 professional university colleges. In this table United States of America, Brazil, India, China and Russian Federation are at the top of the list. Google has been used for creating map of the world with separated countries. Countries’ data which are collected at January 2014 have been saved in private MS Access database. Each country’s name has a unique two character which is standard in the world. Traffic data of the all countries and the two characters code have been sent to Google for creating the color map of the countries. Figure (1) has been created depend on real data of the all countries based on number of universities in each country.
Table 1:
Top 60 countries with highest number of universities
Rank
Country Name Number of Universities
Rank
Country Name Number of Universities
Rank
Country Name Number of Universities
1 United States of America 3344 21 Italy 225 41 Malaysia 82 2 Brazil 1834 22 Thailand 183 42 Hungary 82 3 India 1743 23 Turkey 170 43 Austria 77 4 China 1252 24 Taiwan 170 44 Georgia 77 5 Russian Federation 1088 25 Netherlands 156 45 Bosnia and Herzegovina 75 6 Mexico 962 26 Nigeria 144 46 Algeria 74 7 Japan 861 27 Vietnam 124 47 Ecuador 72 8 France 635 28 Kazakstan 120 48 Venezuela 71 9 Iran (Islamic Republic of Iran) 605 29 Portugal 118 49 Norway 67 10 Poland 475 30 Argentina 117 50 Greece 67 11 Germany 425 31 Switzerland 113 51 Saudi Arabia 62 12 Republic Of Korea 419 32 Romania 111 52 Egypt 60 13 Indonesia 373 33 Bangladesh 107 53 Costa Rica 59 14 Pakistan 344 34 Morocco 105 54 Bulgaria 59 15 Ukraine 336 35 Australia 104 55 Uzbekistan 58 16 United Kingdom 330 36 Belgium 99 56 Latvia 58 17 Philippines 307 37 Denmark 98 57 Finland 57 18 Colombia 306 38 Peru 92 58 Belarus 56 19 Canada 265 39 Czech Republic 85 59 Tunisia 55 20 Spain 248 40 Chile 85 60 Iraq 55
Figure 1:
Countries are colored based on number of universities in each country
Table (2) shows the top 80 universities and their associated weights which is taken in January 2014.
Table 2:
Weight of top 80 universities
Rank University Name 𝐖 𝐮 Rank University Name 𝐖 𝐮
1 Singapore-MIT Alliance for Research and Technology 0.000864 41 Yale University 0.000214 2 Harvard-MIT Division of Health Sciences and Tecnology 0.000858 42 Carnegie Mellon University 0.000207 3 Massachusetts Institute of Technology 0.000852 43 Centro Universitario Estadual da Zona Oeste 0.000197 4 Stanford University 0.000714 44 University of Florida 0.000197 5 Harvard University 0.000644 45 University of Oxford 0.000189 6 Universidad Nacional Mexico 0.000509 46 University of Applied Science and Technology Tehran 0.000178 7 University of California Berkeley 0.000463 47 University of Toronto 0.000175 8 Pennsylvania State University 0.000419 48 Victoria University in the University of Toronto 0.000174 9 Columbia University New York 0.000373 49 Trinity College in the University of Toronto 0.000174 10 Cornell University 0.000346 50 University of Southern California 0.000171 11 Weill Medical College Cornell University 0.000341 51 Harvard University Harvard Business School 0.000163 12 Weill Cornell Medical College in Qatar 0.000340 52 Ohio State University 0.000161 13 University of Texas Austin 0.000314 53 University of British Columbia 0.000161 14 University of Michigan 0.000302 54 University of California Davis 0.000156 15 New York University 0.000300 55 Payam Noor University Kabodrahang 0.000151 16 University of Michigan Dearborn 0.000298 56 Payam Noor University 0.000150 17 University of Wisconsin Madison 0.000258 57 University of Cambridge 0.000149 18 University of Pennsylvania 0.000253 58 University of California San Diego 0.000148 19 University of Minnesota 0.000251 59 Rutgers University 0.000148 20 University of Washington 0.000250 60 University of Phoenix 0.000147 21 University of Minnesota Duluth 0.000249 61 Universidade Paulo USP 0.000147 22 University of Minnesota Morris 0.000247 62 Michigan State University 0.000147 23 University of Minnesota Crookston 0.000247 63 Academy of State Fire Service 0.000146 24 University of Minnesota, Rochester 0.000246 64 Rutgers University Camden 0.000144 25 University of Illinois Urbana Champaign 0.000246 65 University of Tehran 0.000143 26 University of California Los Angeles UCLA 0.000244 66 Palawan State University 0.000139 27 Purdue University 0.000243 67 Universidad Santo Tom 0.000138 28 Princeton University 0.000243 68 Escuela de Arquitectura de Chihuahua 0.000138 29 Moscow Regional Social and Economic Institute 0.000243 69 University of Maryland 0.000137 30 CUNY Medgar Evers College 0.000217 70 University of North Carolina Chapel Hill 0.000135 31 CUNY John Jay College of Criminal Justice 0.000217 71 North Carolina State University 0.000133 32 City University of New York 0.000217 72 Arizona State University 0.000133 33 CUNY York College 0.000217 73 College of Law Latvia 0.000132 34 CUNY New York City College of Technology 0.000217 74 Boston University 0.000132 35 CUNY Queens College 0.000217 75 Escola de Governo Professor Paulo Neves de Carvalho 0.000132 36 CUNY Brooklyn College 0.000217 76 Helena Antipoff FHA 0.000132 37 CUNY Hunter College 0.000216 77 Trabalho de Minas Gerais UTRAMIG 0.000132 38 City College of New York CUNY 0.000216 78 University of Anatolia 0.000130 39 CUNY College of Staten Island 0.000216 79 Duke University 0.000129 40 CUNY Baruch College 0.000216 80 University of Arizona 0.000127
Table (3) shows top 60 countries with highest calculated weights. United State of America has gained 38.6 percent of academic traffic of the world. Although, countries with more universities have higher chance to have better rank, but, having more universities doesn’t guarantee to take the better rank, for example Russian Federation which is the 5’th country in table (1), sits in the 11’th position in table (3). igure (2) shows colored countries which are based on weights of each country. In figure (2) the color of each country are selected based on weight of academic traffic in table (3).
Table : Top 60 countries with highest weight of academic traffic
Rank
Country Name Number of Universities
Weight of Country ( W c ) Rank
Country Name Number of Universities
Weight of Country ( W c )
1 United States of America
31 Chile
2 India
32 Republic Of Korea
3 Brazil
33 Malaysia
4 China
34 Greece
5 Iran (Islamic Republic of Iran)
35 Belgium
6 United Kingdom
36 Austria
7 Canada
37 Peru
8 Germany
38 Finland
9 Spain
39 Singapore
10 France
40 Czech Republic
11 Russian Federation
41 South Africa
12 Japan
42 Portugal
13 Mexico
43 Norway
14 Australia
44 Vietnam
15 Italy
45 Philippines
16 Indonesia
46 Hong Kong
17 Turkey
47 Bangladesh
18 Taiwan
48 Kazakstan
19 international
49 Nigeria
20 Poland
50 Ukraine
21 Saudi Arabia
51 Denmark
22 Colombia
52 Egypt
23 Cuba
53 Ireland
24 Netherlands
54 Venezuela
25 Pakistan
55 Latvia
26 Qatar
56 Hungary
27 Thailand
57 Bolivia
28 Switzerland
58 New Zealand
29 Sweden
59 Romania
30 Argentina
60 Belarus Figure 2:
Countries are colored based on traffic’s weight of each country
Table 4:
Average weight of all universities in each country
Country Name
Country’s Weight
Number of Universities Average Weight of Country ( A wc ) ∗ 10 Australia
Canada
United States of America
United Kingdom
Spain
Germany
Turkey
Iran (Islamic Republic of Iran)
Italy
Taiwan
Switzerland
Argentina
Netherlands
France
Indonesia
China
India
Thailand
Portugal
Vietnam
Bangladesh
Japan
Brazil
Colombia
Kazakstan
Mexico
Russian Federation
Poland ountry Name
Country’s Weight
Number of Universities Average Weight of Country ( A wc ) ∗ 10 Pakistan
Nigeria
Romania
Republic Of Korea
Philippines
Ukraine
Morocco
By substituting the real values in above formula the total weights of universities’ sites equal to 0.0890988, the total weights of all sites is equal to 17.7939 and P 𝑎𝑡 is equal to 0.50072 %. P 𝑎𝑡 =0.5% shows academic traffic of the 21,485 academic sites in the 30 million sites. The results show that about 0.5 percent of all traffics of the World Wide Web belong to academic traffic. One hit of each 200 hits in the internet, belongs to academic sites. 21,485 sites are 0.073 percent of all Alexa’s sites (by dividing 21,485 to 30 millions) but 0.5 percent of traffics belong to academic web sites. By dividing 0.5 to 0.073 we will reach to 6.85, this means average of hitting academic web sites are 6.85 times of average hits per all sites around the world. United States of America, India, Brazil, China and Iran (Islamic Republic of Iran) are at the top of the list of countries which use most academic traffic of the world. 38.6 percent of academic traffic belongs to United State of America. Australia, Canada, United States of America, United Kingdom and Spain are 5 countries with highest average weight of countries’ academic traffic. Discussion
Rankings of counties based on academic traffic have been investigated in this study. As mentioned in the previous sections, academic traffic rank of countries has been calculated based on universities traffic rank. In this case, countries with higher number of universities have been more chance to gain more weight. The selected approach hides important information. Two university sites close to each other regarding ranks might be far from each other with respect to number of hits. Conversely, two university sites far from each other regarding ranks might be close to each other with respect to number of hits. For overcoming this problem we have used highest number of universities. In this case, differences between ranks are at minimum quantity then two university sites close to each other regarding ranks are close to each other with respect to number of hits. More universities don’t guarantee to gain better rank, for example Russian Federation which is the 5’th country in table (1), sits in the 11’th position in table (3). In the other hand some countries with lower number of universities, sit in higher rank in table (3), for example Iran which is the 9’th country in table (1), sits in the 5’th position in table (3). Now, the main question is: What are the most important indicators for countries to take better academic traffic rank? The number of population, number of universities and performance of sites are some candidates which have been more considered to study in this section. ist of countries and their populations are presented in table (5) and map of the world population is presented in right part of figure (3) based on World Bank, which shows a relation between number of countries’ population and number of universities in each country.
Table 5:
List of countries by population (http://data.worldbank.org/indicator/SP.POP.TOTL?order=wbapi_data_value_2013+wbapi_data_value+wbapi_data_value-last&sort=desc) Country name
1 China
2 India
3 United States
4 Indonesia
5 Brazil
6 Pakistan
7 Nigeria
8 Bangladesh
9 Russian Federation
10 Japan
11 Mexico
12 Philippines
13 Ethiopia
14 Vietnam
15 Egypt, Arab Rep.
16 Germany
17 Iran, Islamic Rep.
18 Turkey
19 Congo, Dem. Rep.
20 Thailand
21 France
22 United Kingdom
23 Italy
24 Myanmar
25 South Africa
26 Korea, Rep.
27 Tanzania
28 Colombia
29 Spain
30 Ukraine
Map of the world population is shown in right part of figure (3). Left map are countries which have been colored based on number of universities in each country.
Figure 3:
Map of the world population. Legend: 0 to 50 M to 400 M to 1,336 M (million), 2011: Left map has been created with Google based on our database and right map has been copied from the World Bank.
The second candidate is number of universities. Normally, there is a relation between counties’ population and number of universities. More population lead to more universities. Each university has its members who only visit its university’s site. These members don’t visit other sites, then more number of universities don’t guarantee to gain better rank. More Population forces to more universities but, doesn’t make more visitors for all universities of the country. Another indicator which has been investigated in this study is language of the countries. Figure (4) shows the percentage of English speakers by countries (right map). This map is more similar to countries which have been colored based on traffic’s weight of each country (left map).
Figure 4:
Weight of countries’ academic traffic (left map) and percentage of English speakers by country (right map.
Left map has been created with Google based on our database and right map has been taken from Wikipedia http://en.wikipedia.org/wiki/List_of_countries_by_English-speaking_population) \80-100% 60-80% 40-60% 20-40% 0-20%
There is a relation between countries’ academic traffic and language of sites in each country. Four top countries with higher average of academic traffic in the world speak in English (table (4)). Conclusion & future work eight of university’s traffic could be a good parameter to estimate percentage of universities’ real traffic. We can use Weight of country’s traffic (W c ) for comparing academic traffic of the countries. New indicator percentage of academic traffic (P 𝑎𝑡 ) is introduced based on weight of university’s traffic (W u ). It is recommended to investigate relation between country’s academic traffic and speed of internet in the countries. We guess that more internet speed lead to more click. Another good subject to investigate is comparing changes of universities traffic rank in each 2-3 months. Comparing these ranks lead to scholars to investigate effect of registration period of universities and effect of vacations in their Alexa’s rank.
References ِ rneborn, L., & Ingwersen, P. (2004). Toward a basic framework for Webometrics. Journal of the American Society for Information Science and Technology, 55(14), 1216–1227. Bordons, M., Fernandez, M.T., & G mez, I. (2002). Advantages and limitations in the use of impact factor measures for the assessment of research performance in a peripheral country. Scientometrics, 53(2), 195–206. Cronin, Blaise. "Bibliometrics and beyond: some thoughts on web-based citation analysis." Journal of Information Science 27.1 (2001): 1-7. Cronin, Blaise, Elisabeth Davenport, and TAYLOR GRAHAM. "Post-professionalism: Transforming the information heartland." (2013). Ding, Ying, and Blaise Cronin. "Popular and/or prestigious? Measures of scholarly esteem." Information processing & management 47.1 (2011): 80-96. Eccles, C. (2002). The use of university rankings in the United Kingdom. Higher Education in Europe, 27(4), 423–432. Egghe, L. (2000). New informetric aspects of the Internet: Some reflections, many problems. Journal of Information Science, 26(5), 329–335. Ghane, M. R., Khosrowjerdi, M., & Azizkhani, Z. (2013). The ranking of Iranian universities based on an improved technique. Malaysian Journal of Library & Information Science, 18(2), 33-45. . uang, C., & Chang, S. (2009). Commonality of web site visiting among countries. Journal of the American Society for Information Science and Technology, 60(6), 1168–1177. Huang, M.-H., Lin, C.-S., & Chen, D.-Z. (2011). Counting methods, country rank changes, and counting inflation in the assessment of national research productivity and impact. Journal of the American Society for Information Science and Technology, 62(12), 2427–2436. Lee, J. Min, J. Oh, A. & Chung, C. (2014). Effective ranking and search techniques for Web resources considering semantic relationships. Information Processing & Management, 50(1), 132-155. Li, J., Sanderson, M., Willett, P., & Norris, M. (2010). Ranking of library and information science researchers: Comparison of data sources for correlating citation data, and expert judgments. Journal of Informetrics, 4(4), 554–563. Li, Y., Castellano, C., Radicchi, F., & Ruiz-Castillo, J. (2013). Quantitative evaluation of alternative field normalization procedures. Journal of Informetrics, 7(1), 746-755. Lin, C. C., & Chien, T. K. (2014). The Effects of Popularity: An Online Store Perspective. International Journal of Information Science and Management (IJISM) , 1-11. Lin, C., Huang, M., & Chen, D. (2013). The influences of counting methods on university rankings based on paper count and citation count. Journal of Informetrics, 7(3), 611–621. Liu, N. C., & Cheng, Y. (2005). The academic ranking of world universities methodologies and problems. Higher Education in Europe. 30(2), 127–136. Marginson, S., & van der Wende, M. (2007). To rank or to be ranked: The impact of global rankings in higher education. Journal of Studies in International Education, 11(3/4), 306–329. Norris, M., &Openheim, C. (2007). Comparing alternatives to the Web of Science for coverage of the social sciences literature. Journal of Informetrics, 1(2), 161–169. Ortega, J.L. and Aguillo, I.F. (2008b), Visualization of the Nordic academic web: link analysis using social network tools, Information Processing and Management, Vol. 44 No. 4, pp. 1624-33. Ortega, J.L. and Aguillo, I.F. (2010), Differences between web sessions according to the origin of their visits, Journal of Informetrics. 4 (3), 331-337.
Peeters, H., Verschraegen, G., & Debels, A. (2014). Commensuration and policy comparison: How the use of standardized indicators affects the rankings of pension systems.
Journal of European Social Policy , (1), 19-38. Regional information center for science and technology. (2013). About. Retrieved from: http://en.ricest.ac.ir/ chubert, A., Glonzel, W., & Braun, T. (1987). A new methodology for ranking scientific institutions. Scientometrics, 12, 267-292. Schubert, A., Glonzel, W., & Braun, T. (1987). A new methodology for ranking scientific institutions. Scientometrics, 12, 267-292. Thelwall, M., & Vaughan, L. (2004). Webometrics: An introduction to the special issue. Journal of the American Society for Information Science and Technology, 55(14), 1213–1215. Thelwall, M., Vaughan, L., & Bj ِِ