Can Cultural Indices Affect Innovation
Muralidhar M.A.
Thursday, August 07, 2014
## Loading required package: cluster
## Loading required package: grid
## Loading required package: lattice
## Loading required package: survival
## Loading required package: splines
## Loading required package: Formula
## Hmisc library by Frank E Harrell Jr
##
## Type library(help='Hmisc'), ?Overview, or ?Hmisc.Overview')
## to see overall documentation.
##
##
## Attaching package: 'Hmisc'
##
## The following object is masked from 'package:survival':
##
## untangle.specials
##
## The following object is masked from 'package:car':
##
## recode
##
## The following objects are masked from 'package:base':
##
## format.pval, round.POSIXt, trunc.POSIXt, unitsIntroduction
This analysis seeks to answer the question: why is it that select countries produce Nobel laureates whereas other countries do not. The answers are based on the analysis of the following parameters:
Research expenses allocated by a country as a percentage of its GDP
Number of researchers per 1 Million people living in a country
Number of articles published per Million People in a country
PDI, IDV factors from Hofstede’s cultural dimensions theory[1].
Hofstede defines PDI and IDV:
PDI: “Power Distance Index measures the extent to which the less powerful members of organizations and institutions (like the family) accept and expect that power is distributed unequally.” (e.g. Malaysia - 104 vs Austria - 11)
IDV: “Individualism is the one side versus its opposite, collectivism, that is the degree to which individuals are integrated into groups. On the individualist side we find societies in which the ties between individuals are loose: everyone is expected to look after him/herself and his/her immediate family. On the collectivist side, we find societies in which people from birth on wards are integrated into strong, cohesive in-groups, often extended families (with uncles, aunts and grandparents) which continue protecting them in exchange for unquestioning loyalty.” (e.g.United States - 91 vs Guatemala - 6)
Methods
Data Collection:
Data required for this analysis were obtained from the following listed websites which are available to public.
List of countries by Nobel laureates per capital was sourced from Wikipedia [2]. List includes prizes awarded for the following disciplines: Chemistry, Literature, Peace, Physics, Physiology or Medicine and Nobel Memorial Prize in Economic Sciences and includes all Nobel Prizes awarded up to and including 14 October 2013. Number of Laureates per 10 million people was calculated based on countries population in the year 2013.
Researchers in R&D (per million people) and Research and development expenditure (% of GDP) for the year 2009 were extracted from a dataset “WDI-Data”, downloaded from website of World Bank[3].
National cultures, Power distance Index (PDI) and Individualism(IDV) were sourced from the website: Clearly Cultural - Making Sense of Cross Cultural Communication[1]. These values are based on the analysis of Geert Hofstede’s cultural dimensions. These indices do not indicate a period of measurement because the evolution of a country’s culture typically takes place over a long period of time.
This Analysis assumes it takes several years of research work and publications to be ever nominated for a Nobel Prize. Hence choosing to consider the year 2009 for Researchers in R&D (per million people) and Research and development expenditure (% of GDP) is acceptable for the analysis of the number of Nobel Prizes awarded in 2013.
Exploratory Analysis:
Exploratory analysis was performed by examining tables and plots (Figure (1)) of the observed data; identified transformations on the raw data on the basis of plots and knowledge of the scale of measured variables. Exploratory analysis was used to (1) identify missing values, (2) verify the quality of the data, and (3) determine the terms used in the regression model relating number of laureates per capita to research expense as percentage of GDP, researchers per capita and cultural indexes:PDI and IDV.
Statistical Analysis:
This Analysis, a standard multivariate linear regression model was developed to understand how the following factors:
researcherPerMillionPeople (Researchers in R&D (per million people)) in the year 2009: “Researchers in R&D are professionals engaged in the conception or creation of new knowledge, products, processes, methods, or systems and in the management of the projects concerned. Postgraduate PhD students (ISCED97 level 6) engaged in R&D are included” as defined in [4].
researchExpensePerofGDP (Research and development expenditure (% of GDP)) in the year 2009: “Expenditures for research and development are current and capital expenditures (both public and private) on creative work undertaken systematically to increase knowledge, including knowledge of humanity, culture, and society, and the use of knowledge for new applications. R&D covers basic research, applied research, and experimental development” as defined in [5].
journalsPerMillionPeople in the year 2009: Computed here for the analysis using the figures “Scientific and technical journal articles” and “Population,Total” from World Bank Data.
“Scientific and technical journal articles refer to the number of scientific and engineering articles published in the following fields: physics, biology, chemistry, mathematics, clinical medicine, bio medical research, engineering and technology, and earth and space sciences”[6].
“Total population is based on the de facto definition of population, which counts all residents regardless of legal status or citizenship–except for refugees not permanently settled in the country of asylum, who are generally considered part of the population of their country of origin. The values shown are midyear estimates” [7].
- PDI: Power Distance Index as defined by Hofstede’s cultural dimensions theory[1]
IDV: Individualism as defined by Hofstede’s cultural dimensions theory[1]
affect the the Number of Nobel Laureates countries produce. This Analysis assumes the Number of Nobel Laureates from a country is a dependent variable and is transformed into “Laureate10Million”[2] and it depends on dependent variables listed earlier.
Results
This analysis fits Multivariate Regression Model to the following parameters:
- Research expenses allocated by a country as a percentage of its GDP
- Number of researchers per 1 Million people living in a country,
- Number of journal articles published per Million People
- PDI: Power Distance Index
- IDV: Individualism
to predict Nobel Laureates Per 10 Million People(NLPM).
Regression Analysis revealed P values are statistically significant to reject Null Hypothesis of variables Researchers/Million (RPM)and PDI. Final regression model is: \[ NLPM = \beta_0 + \beta_1 RPM + \beta_2 PDI + \epsilon \] where \(\beta_0\) is an intercept term and \(\beta_1\) represents the change in NLPM associated with a change in one unit researchers per 1 Million people while Power distance Index is unchanged. And \(\beta_2\) is the change in NLPM associated with a one unit change in PDI and while researchers per 1 Million people is unchanged.
It was observed that there exist a strong positive relation between NLPM and RPM with \(\beta_1 = 0.0015\) and \(P = 1.5e-02\), as opposed to a strong inverse relation between NLPM and PDI with \(\beta_2 = -0.221\) and \(P = 8.6e-04\). Since P values observed here were very significant, Null Hypothesis were accepted . The effect of PDI is stronger in comparison to RPM based on values of \(\beta\) coefficients .
\(\beta_0\): 13.49 Nobel Laureates Per 10 Million People and this value can vary within 95% confidence interval:6.4 ~ 20.5 Nobel Laureates Per 10 Million.
\(\beta_1\): +0.001 means one researcher Per Million People increase will result in +0.001 increase in Nobel Laureates Per 10 Million People and associated with 95% confidence interval:+0.0004 ~ +0.002 Laureates Per 10 Million People for every one researcher Per Million People.
\(\beta_2\): -0.19 means one init of PDI increase will result in -0.19 decrease in Nobel Laureates Per 10 Million People and associated with 95% confidence interval:-0.28 ~ -0.1 Laureates Per 10 Million People for every unit of PDI.
Figure 1, shows residuals bounce randomly around the 0 line. Hence the assumption considered in this relationship is linear and reasonable. 
During the development of the model, Null Hypothesis was accepted for the following independent variables:
- Research expenses allocated by a country as a percentage of its GDP
- Number of articles published per Million People in a country
since P values were greater than 0.05. Inspite of the lack of influence on Number of Nobel Laureates produced from a country, it is of interest to observe the subtle influence of these variables through faceting in Figures (3) & (4).
In, Figure (3) and (4), matrices of graphs were created by splitting data into subsets which fell in 4 quantiles obtained from
- Research expenses allocated by a country as a percentage of its GDP
- Number of articles published per Million People in a country
Table-1 shows, range of values values assigned to each quartiles.
Table-1
| Quantile | 25% | 50% | 75% | 100% |
|---|---|---|---|---|
| Research Expense/GDP | Rech. Exp/GDP-Q1 | Rech. Exp/GDP-Q2 | Rech. Exp/GDP-Q3 | Rech. Exp/GDP-Q4 |
| Range (%) | 0.0 ~ 0.05 | 0.06 ~ 0.45 | 0.46 ~ 1.26 | 1.27 ~ 1.87 |
| — | — | – | — | —- |
| Articles/Million People | Jrnl/Million-Q1 | Jrnl/Million-Q2 | Jrnl/Million-Q3 | Jrnl/Million-4 |
| Range (count) | 0 ~ 1.1 | 1.2 ~ 53.87 | 53.88 ~ 216.11 | 216.12 ~ 630.24 |
In Figure (3), the countries in the upper left quarter, it was observed that they employed less than 1000 Researchers/Million with allocation of research expenditure less than 0.45% of their GDP and published less than 216 journal articles/Million. The number of Noble Prizes won in these countries was found to be 1 per 10 Million people. However, Hungary is an outlier for winning more than 9 prizes which can be attributed to their PDI being less than 50 (Figure (4)).
The countries in the lower right quarter, it was observed that they employed more than 1000 Researchers/Million with allocation of research expenditure less than 0.45% of their GDP and published more than 216 journal articles/Million. The number of Nobel Prizes won in these countries found to be 9 per 10 Million people. 
Italy, Czech Republic, Spain, Portugal, Japan and Singapore are outliers for winning less than 5 Nobel prizes in spite of engaging more than 2000 Researchers/Million People with allocation fund exceeding 0.5% of their GDP can be attributed to their PDI is greater than 50 (Figure (4)).
Table 2 shows prediction from the model developed here to the actual values of Nobel Laureates per 10 Million. Though the predictions are in the same order of magnitude, it requires data for more countries to improve this model.
Table-2
| Country | Researchers/Mil | PDI | Model Prediction | Actual |
|---|---|---|---|---|
| USA | 4042 | 39 | 11.54 | 10.8 |
| UK | 4134 | 35 | 12.4 | 19.65 |
| India | 160 | 77 | -0.84 | 0.072 |
| Turkey | 900 | 60 | 3.63 | 0.13 |
References
- Clearly Cultural- http://www.clearlycultural.com/
- List of countries by Nobel Laureates - http://en.wikipedia.org/wiki/List_of_countries_by_Nobel_laureates_per_capita
- 5.13 World Development Indicators:Science and technology - http://wdi.worldbank.org/table/5.13
- Researchers in R&D (per million people) http://data.worldbank.org/indicator/SP.POP.SCIE.RD.P6
- Research and development expenditure (% of GDP) http://data.worldbank.org/indicator/GB.XPD.RSDV.GD.ZS
- Scientific and technical journal articles http://data.worldbank.org/indicator/IP.JRN.ARTC.SC
- Population, Total http://data.worldbank.org/indicator/SP.POP.TOTL
- R Markdown - Dynamic Documents for R http://rmarkdown.rstudio.com/?version=0.98.953&mode=desktop