2016 Ranking Adjusted for Economic Development
Methodology of adjusting for levels of economic development
In order to adjust for national levels of income we regress the values for each variable, in original units, on a function of GDP per capita using data for all 50 countries. The GDP we use is for 2013 in US dollars measured in Purchasing Power Parity (PPP) terms. Both linear and quadratic relationships are used. The fitted equation gives the expected value of a variable for a nation’s level of income. The difference between the actual and expected value will be positive or negative depending on whether a country performs above or below the expected value. Where data are missing we assume that the variable takes the expected value for that country’s level of GDP per head, that is, we assume a deviation value of zero. For the two Output variables based on the Shanghai rankings (O4 and O5) the presence of zero values limits the use of regression, so instead we rank the countries by GDP per head and take a moving average of actual scores to derive more robust estimates of predicted values.
In aggregating over variables we express deviations from the regression line as a percentage of the average of the actual and predicted values and then sum. To use the percentage deviations from the line would ignore the fact that the predicted values below the line are capped at 100 per cent whereas there is no limit above the line. Our method ensures symmetry in that values that are half what is expected at a given level of GDP per capita have the same influence as values that are double those expected. By construction, our calculated deviations lie in the range –200 per cent to + 200 per cent. The average deviation for each module is a weighted sum of the deviations for each of the measures within the module. The method of measuring deviations needs to be borne in mind when interpreting the weighted average numerical scores for each module and for the overall ranking.
We use the same dependent variables and weights as described in section 3 with two exceptions. The exceptions are research expenditure (R4 and R5) and publication output (O1 and O2) where in each case we had variables expressed in two different forms. This becomes unnecessary when we control for differences in income levels. We delete R5 and move the weight to R4, so that each of the four resources measures has a weight of 5 per cent in the overall ranking. In the output module we use as a single publication measure the number of articles divided by (total) GDP, thus combining O1 and O2.
Expenditure levels are best described by a linear relationship with GDP except for research expenditure where a quadratic curve fits best. The top two countries devoting greater resources to higher education than is expected at their level of GDP per capita are Serbia and Malaysia, where the scores are 40 per cent above expected. Next, in order, are Denmark, Ukraine, India, Canada and Sweden, Finland and South Africa, where scores exceed 20 per cent above expected. The high values for Serbia and the Ukraine are at least partly explained by the combination of ‘sticky’ expenditure on higher education and falling GDP per capita, but the deviations from expected expenditure have declined significantly from last year’s rankings. Compared with the non-adjusted rankings, China, India and South Africa each improve 33 places, China to 12thposition.
Turning to the four variables that are included in the Resources module, total expenditure on higher education as a share of GDP shows only a very slight increase with GDP per head so the rankings are similar to those discussed in section 3.1. The highest ranked countries are Chile, Malaysia and the United States. Government expenditure shows a little more variation but the relationship is still quantitatively small: for every ten thousand dollar increase in GDP per head, government expenditure is estimated to increase by around 0.1 of a percentage point. The top countries for government expenditure after adjusting for GDP per capita are Ukraine, Saudi Arabia, India and Malaysia. Expenditure (which includes research expenditure) per student increases markedly with income levels: on average by around USD400 for each USD1,000 increase in GDP per capita (adjusted R2= 0.759). The top eight countries on an income-adjusted basis are, in order, South Africa, Malaysia, the United Kingdom, Brazil (public institutions only), India, China, Sweden and the United States.
Research expenditure in higher education as a share of GDP increases with GDP per capita, but at a declining rate. The quadratic regression estimates imply that at GDP per head of USD 25,000 the expected expenditure on R&D is 0.35 per cent of GDP whereas the corresponding figure for GDP per head of USD 50,000 is 0.60. The top eight countries for research expenditure as a share of GDP are now Serbia, South Africa, China, Denmark, Turkey, Sweden, Portugal and Switzerland. The United States is ranked at 41, compared with 20th when no allowance is made for income levels, which largely accounts for the United States relatively low overall ranking for resources of 16th.
In principle, the creation of a favourable environment is independent of income levels so we do not carry out regression analysis. Instead, we deviate values from the mean level for each of the five components. In practice only the WEF survey results (E5) show a significant variation with GDP per capita. In order to be consistent with the treatment used in other modules this year we express the deviations from average as a ratio of the mean of actual plus expected (the average). The rankings are necessarily very similar to those for the unadjusted data. For missing data the deviations are put at zero.
The range of scores is lower for this module than for the other three modules because several variables have low variation across countries. The scores for the top four countries (the United States, Hong Kong SAR, New Zealand and Finland) are around 10 per cent above expected values.
The connectivity measures are quite strongly positively related to levels of GDP per head, with the adjusted R2 values lying in the range 0.3 to 0.5. However, even after allowing for income differences between countries, of the top five ranked nations, four are developed countries: the United Kingdom, New Zealand, Switzerland and Denmark. The highest ranked low-income country is South Africa, ranked at number three. The next highest ranked developing countries are Thailand (10th) and Indonesia (14th). South Africa ranks very highly in all but the Web-based measures of connectivity.
The strongest relationship with income levels is obtained for the ranking of ‘knowledge transfer between industry and universities’ (C5). Of course causality can run both ways: do high levels of connectivity between industry and higher education institutions raise GDP per head or do high levels of income enable universities to concentrate more on research and development. Malaysia and Israel top the income-adjusted performance level for knowledge transfer. The equation for international co-authorship (C2) implies that for every USD1,000 increase in GDP per capita the percentage of articles that have an international co-editor increase by 3.2 percentage points. The top three ranked countries are South Africa, Chile and Indonesia.
Joint authorship with industry (C6) flattens out at high income levels and is best approximated by a quadratic relationship. India and Indonesia perform well relative to their levels of GDP per head. After allowing for income differences China ranks in the top ten for Web impact (C4) and links with industry (C5, C6).
All but one of the Output measures show a significant increase with levels of GDP per capita. The exception is the unemployment variable (C8) where we deviate from the average value. Two Output measures show a particularly strong relationship with GDP per capita (adjusted R2 > 0.6): impact as measured by citations (O3) and researchers per head of population (O8). The impact measure picks up not only the quality of research but its nature: applied research in developing countries is unlikely to be highly referenced despite its relevance for economic development. To illustrate, after allowing for income differences, Malaysia is ranked first for knowledge transfer (C5) but ranks only 35th for impact through citations. While both the number of researchers per head and tertiary enrolment rates increase with income levels they flatten out at the highest income levels.
The top three ranked countries for Output are Serbia, China and Israel, all nearly 40 per cent above expected values for their level of income. Next in order come the United Kingdom and Portugal. Turning to the components, Serbia, India and Portugal are the top three for publications relative to (total) GDP. India and South Africa do best for the average impact of publications (at around 40 per cent above the expected value) followed by the United Kingdom, Italy, the Netherlands and Indonesia. Relative to income levels, the quality of the best three universities is highest in China and the United States. Next in order are the United Kingdom, Brazil, Israel and South Africa.
After allowing for income levels, Ukraine is ranked first on participation rates (O6), qualification of the workforce (O7) and number of researchers (O8). Israel is third for both the qualification of the workforce and numbers of researchers. The countries ranked second are Russia (for qualification of the workforce) and China (for number of researchers).
The overall score is calculated by weighting the percentage deviations for each module. The weights are the same as for the unadjusted data: Resources (20%), Environment (20%), Connectivity (20%) and Output (40%). The aggregate percentage absolute scores are only indicative of absolute performance. The median aggregate score is minus eight per cent so that a score above this level can be interpreted as above average. Even this interpretation is dependent on our choice of 50 countries.
The top five countries in rank order are the United Kingdom, Serbia, Denmark, Sweden and China. China has improved from 16thin the 2016 rankings, even though its income levels are rising at above average rates that would otherwise lower the ranking. Of the other countries with per capita GDP below USD20,000 (PPP), South Africa is ranked seventh and India 15th . Malaysia and Brazil are above the median value but Thailand and Indonesia are in the bottom decile.
Compared with the original rankings, six countries are now ranked at least fifteen places higher. These countries are, in order of the ranking improvement, Serbia, India, South Africa, China, Portugal and Brazil.