Robust measurement of innovation performances in Europe with a hierarchy of interacting composite indicators

Figures & data

Figure 1. Example on an hierarchy of criteria with 10 elementary criteria taken from Angilella et al. (2018).

Figure 2. Flow chart of the applied methodology.

Figure 3. Hierarchical structure framework for the EIS.

Table 1. EIS description for the elementary criteria.

		Description of the elementary criteria descending from the macrocriteria HR, ARS, IFE, etc…
FC ($g_1$)	HR ($g_{(1,1)}$)	New doctorate graduates per 1000 population aged 25–34 ($g_{(1,1,1)}$)
		Percentage population aged 25–34 having completed tertiary education ($g_{(1,1,2)}$)
		Percentage population aged 25–64 participating in lifelong learning ($g_{(1,1,3)}$)
	ARS ($g_{(1,2)}$)	International scientific co-publications per million population ($g_{(1,2,1)}$)
		Scientific publications among the top-10% most cited publications worldwide as percentage
		of total scientific publications of the country ($g_{(1,2,2)}$)
		Foreign doctorate students as a percentage of all doctorate students ($g_{(1,2,3)}$)
	IFE ($g_{(1,3)}$)	Broadband penetration: Percentage of enterprises with a maximum contracted download
		speed of the fastest fixed internet connection of at least 100 Mb/s ($g_{(1,3,1)}$)
		Opportunity-driven entrepreneurship: ratio between the share of persons
		involved in improvement-driven entrepreneurship and the share of
		persons involved in necessity-driven entrepreneurship ($g_{(1,3,2)}$)
IN ($g_2$)	FS ($g_{(2,1)}$)	R&D expenditure in the public sector (percentage of GDP) ($g_{(2,1,1)}$)
		Venture capital investments (percentage of GDP) $\left(g_{(2,1,2)}\right)$
	FI ($g_{(2,2)}$)	R&D expenditure in the business sector (percentage of GDP) $\left(g_{(2,2,1)}\right)$
		Non-R&D innovation expenditure (percentage of turnover) $\left(g_{(2,2,2)}\right)$
		Enterprises providing training to develop or upgrade ICT skills of their personnel $\left(g_{(2,2,3)}\right)$
IA ($g_3$)	IT ($g_{(3,1)}$)	SMEs with product or process innovations ($g_{(3,1,1)}$)
		SMEs with marketing or organizational innovations ($g_{(3,1,2)}$)
		SMEs innovating in-house ($g_{(3,1,3)}$)
	LIN ($g_{(3,2)}$)	Innovative SMEs collaborating with others ($g_{(3,2,1)}$)
		Public-private co-publications ($g_{(3,2,2)}$)
		Private co-funding of public R&D expenditures ($g_{(3,2,3)}$)
	IAS ($g_{(3,3)}$)	PCT patent applications ($g_{(3,3,1)}$)
		Trademark applications ($g_{(3,3,2)}$)
		Design applications ($g_{(3,3,3)}$)
IMP $\left(g_4\right)$	EI $\left(g_{(4,1)}\right)$	Employment in knowledge-intensive activities $\left(g_{(4,1,1)}\right)$
		Employment fast-growing firms innovative sectors $\left(g_{(4,1,2)}\right)$
		Medium and high-tech product exports $\left(g_{(4,2,1)}\right)$
	SE $\left(g_{(4,2)}\right)$	Knowledge-intensive services exports $\left(g_{(4,2,2)}\right)$
		Sales of new-to-market and new-to-firm innovations $\left(g_{(4,2,3)}\right)$

Source: Hollanders, Es-Sadki, and Kanerva (2016).

Table 2. Denmark Rank Acceptability Indices, best and worst positions and three highest rank acceptability indices.

(a) Comprehensive level	Best ($b_{k,0}^{Best}$)	Worst ($b_{k,0}^{Worst}$)	$hig{h}_1$ ($b_{k,0}^{hig{h}_1}$)	$hig{h}_2$ ($b_{k,0}^{hig{h}_2}$)	$hig{h}_3$ ($b_{k,0}^{hig{h}_3}$)
University	1 (26.07%)	2 (73.93%)	2 (73.93%)	1 (26.07%)	28 (−)
Industry	2 (8.74%)	7 (13.02%)	5 (27.38%)	3 (21.84%)	6 (17.69%)
Government	2 (22.77%)	6 (0.29%)	3 (29.89%)	5 (29.08%)	2 (22.77%)
(b) Framework Conditions (FC)	Best ($b_{k,1}^{Best}$)	Worst ($b_{k,1}^{Worst}$)	$hig{h}_1$ ($b_{k,1}^{hig{h}_1}$)	$hig{h}_2$ ($b_{k,1}^{hig{h}_2}$)	$hig{h}_3$ ($b_{k,1}^{hig{h}_3}$)
University	1 (99.99%)	2 (0.01%)	1 (99.99%)	2 (0.01%)	28 (−)
Industry	1 (98.29%)	2 (1.71%)	1 (98.29%)	2 (1.71%)	28 (−)
Government	2 (22.77%)	6 (0.29%)	3 (29.89%)	5 (29.08%)	2 (22.77%)
(c) Investments (IN)	Best ($b_{k,2}^{Best}$)	Worst ($b_{k,2}^{Worst}$)	$hig{h}_1$ ($b_{k,2}^{hig{h}_1}$)	$hig{h}_2$ ($b_{k,2}^{hig{h}_2}$)	$hig{h}_3$ ($b_{k,2}^{hig{h}_3}$)
University	1 (80.26%)	4 (0.25%)	1 (80.26%)	2 (16.51%)	3 (2.99%)
Industry	3 (0.17%)	15 (5.52%)	8 (18.82%)	12 (14.76%)	10 (13.55%)
Government	3 (0.35%)	15 (0.57%)	12 (15.82%)	10 (15.80%)	8 (15.56%)
(d) Innovation Activities (IA)	Best ($b_{k,3}^{Best}$)	Worst ($b_{k,3}^{Worst}$)	$hig{h}_1$ ($b_{k,3}^{hig{h}_1}$)	$hig{h}_2$ ($b_{k,3}^{hig{h}_2}$)	$hig{h}_3$ ($b_{k,3}^{hig{h}_3}$)
University	1 (2.86%)	9 (5.31%)	7 (42.14%)	8 (18.56%)	6 (12.72%)
Industry	1 (3.20%)	10 (1.47%)	7 (62.57%)	6 (10.63%)	8 (8.73%)
Government	1 (4.32%)	11 (0.06%)	7 (41.66%)	8 (17.47%)	6 (9.59%)
(e) Impacts (IMP)	Best ($b_{k,4}^{Best}$)	Worst ($b_{k,4}^{Worst}$)	$hig{h}_1$ ($b_{k,4}^{hig{h}_1}$)	$hig{h}_2$ ($b_{k,4}^{hig{h}_2}$)	$hig{h}_3$ ($b_{k,4}^{hig{h}_3}$)
University	7 (0.70%)	15 (0.31%)	12 (45.04%)	11 (14.91%)	9 (12.46%)
Industry	7 (0.06%)	16 (2.27%)	13 (24.64%)	14 (20.41%)	12 (18.12%)
Government	7 (4.70%)	19 (2.78%)	15 (20.86%)	16 (15.60%)	17 (12.43%)

Table 3. Barycenters of the Móbius parameters vectors giving to Denmark the 3rd and the 15th positions, respectively.

	$g_{(2,1,1)}$	$g_{(2,1,2)}$	$g_{(2,2,1)}$	$g_{(2,2,2)}$	$g_{(2,2,3)}$
Baryc_3	0.1596	0.3843	0.3305	0.0697	0.0558
Baryc_15	0.0612	0.4973	0.0530	0.2490	0.1394

Table 4. Expected ranking of countries from the Triple Helix agents at the comprehensive level and significant discrepancies with the EIS ranking.

Country	University	Industry	Government	EIS
Belgium (BE)	9	7	10	9
Bulgaria (BG)	27	27	27	27
Czech Republic (CZ)	13	14	14	13
Denmark (DK)	2	5 (↓ 3)	3	2
Germany (DE)	5	6	9 (↓ 3)	6
Estonia (EE)	14	12 (↑ 3)	12 (↑ 3)	15
Ireland (IE)	11	10	5 (↑ 5)	10
Grece (EL)	20	25 (↓ 3)	24	22
Spain (ES)	17	19	15	17
France (FR)	10	11	8 (↑ 3)	11
Croatia (HR)	26	24	26	26
Italy (IT)	22	21	22	19
Cyprus (CY)	21	17 (↑ 3)	18	20
Latvia (LV)	24	23	23	24
Lithuania (LT)	16	18	19 (↓ 3)	16
Luxembourg (LU)	8	8	7	8
Hungary (HU)	23	16 (↑ 7)	16	23
Malta (MT)	19	20	21 (↓ 3)	18
Netherland (NL)	4	3	4	4
Austria (AT)	6	9	11 (↓ 4)	7
Poland (PL)	25	26	25	25
Portugal (PT)	15	15	17 (↓ 3)	14
Romania (RO)	28	28	28	28
Slovenia (SI)	12	13	13	12
Slovakia (SK)	18 (↑ 3)	22	20	21
Finland (FI)	3	2	6 (↓ 3)	3
Sweden (SE)	1	1	1	1
United Kingdom (UK)	7	4	2 (↑ 3)	5

Figure 4. EU countries innovative performance expected rankings for the Triple Helix agents.

Angilella, S., P. Catalfo, S. Corrente, G. Giarlotta, S. Greco, and M. Rizzo. 2018. “Robust Sustainable Development Assessment with Composite Indices Aggregating Interacting Dimensions: the Hierarchical-SMAA-Choquet Integral Approach.” Knowledge-Based Systems 158: 136–153.

 Web of Science ®Google Scholar

Hollanders, H., N. Es-Sadki, and M. Kanerva. 2016. European Innovation Scoreboard 2016. Brussels: European Commission.

 Google Scholar