Roud - Innovation statistics-is data indifferent to the complexity of firm strategies
0 likes403 views
The document analyzes the relationship between the sophistication of firms' innovation strategies and their ability to comprehend and accurately fill out innovation surveys. It uses data from a 2015 Russian innovation survey of over 1,300 manufacturing and ICT enterprises. Latent class analysis identifies five clusters of firms that differ in their understanding of survey concepts and precision of data provided. Regression analysis finds that factors like organizational innovation, new product innovation, and use of advanced production processes are associated with better comprehension, while abandoned innovation reduces comprehension. The study aims to understand how firms' innovation competencies impact their ability to participate meaningfully in innovation data collection.
Roud - Innovation statistics-is data indifferent to the complexity of firm strategies
- 1. Institute for Statistical Studies and Economics of Knowledge The value of innovation statistics – is data indifferent to the complexity of firm strategies Vitaliy Roud, Leonid Gokhberg 19 September 2016 Blue Sky III – Ghent
- 2. 2 © National Research University Higher School of Economics, 2015 Outline Hypothesis: link between the sophistication of innovation strategy and the comprehension of the innovation survey questionnaire Data and method Strategies for collecting data Perception of the questionnaires Accuracy of data provision Grading the quality of survey fill-in Testing the link between innovation strategy and quality of data provided
- 3. − Qualified innovation managers of larger enterprises are the best to recognize the core concepts − Informal innovators would face all sorts of difficulties 3 Hypothesis: Competences to fill in the innovation survey questionnaire and the sophistication of innovation strategy • Cognitive testing of innovation studies – companies are not equal in comprehension of innovation survey concepts • Two extrema: • This study: to operationalise the continuum of states between total comprehension and total misunderstanding
- 4. Data 4 Monitoring of Innovation Behaviour of Enterprises • Russian branch of the European Manufacturing Survey (Consortium of 18 research centres coordinated by Fraunhofer ISI) • Original methodology compliant with the Oslo Manual, EU CIS and Russian Innovation Survey • Executed by Higher School of Economics Institute for Statistical Studies and Economics of Knowledge biannually since 2009: http://issek.hse.ru/innoproc/en/ • Round 2015: ~1300 enterprises in Manufacturing and ICT • Personal interviews with the top management, stratified representative sample (firm size, sector) Firm-level data on: • Conventional indicators of innovation • Participation in national innovation surveys
- 5. Methods 5 Latent class analysis: comprehension of the innovation survey concepts Latent class analysis: accuracy of data provision Latent class analysis: strategies of data collection Stage 1: understanding the diversity Stage 2: reducing the dimensions Principle component analysis: comprehension and accuracy of data Multiple choice regression (mlogit): grade of quality and the sophistication of innovation strategy Stage 3: testing the heterogeneity Latent class analysis – grade of the survey participation quality: comprehension and accuracy of data Innovation Strategy Technology level General controls
- 6. Strategies of data collection and provision 6 Accounting dept Economic/ Financial planning depts Top management and technical depts Complex: economic planning and technical depts All types of depts Cluster Size 0.424 0.2037 0.1877 0.1602 0.0244 Departments/specialists involved Accounting 0.9998 0.011 0.0009 0.3812 0.9918 Economic and financial planning departments 0.0254 0.9993 0.0016 0.8204 0.9991 Top management 0.2117 0.2146 0.4258 0.2575 0.9407 Innovation department 0.0039 0.0346 0.1597 0.2011 0.9764 Technological and technical departments 0.0632 0.1347 0.3242 0.6167 0.9987 Marketing 0.0085 0.0001 0.0406 0.3318 0.922 HR 0.0278 0 0 0.2703 0.9616 Other 0 0.0006 0.0226 0.0181 0.0092 (Share of enterprises within the cluster involving the corresponding departments) Enterprise departments involved in data collection: 42% are filled in exclusively by accounting department 0 50 100 150 200 1 2 3 4 5 6 7 8 9 10 12 15 20 23 24 25 30 37 Numberofenterprises Employees involved in data collection
- 7. Perception of the Innovation Survey Concepts (1) 7 Portfolio of questionnaire perceptions: perfectly relevant vs. non-applicable for the firm Perfect applicability Good applicability Average applicability Average for core questions - Poor for extended Poor applicability Cluster Size 0.1529 0.2529 0.3549 0.1194 0.1199 Indicators General firm characteristics (markets, human capital, etc.) 1.0554 1.7768 2.7448 1.8903 4.1099 Innovation types: product, process, organisational, marketing 1.0304 1.9682 3.0167 3.1313 4.9992 Innovation sales 1.0115 1.9536 3.1583 3.2661 4.9887 Factors hampering innovation 1.0657 1.8789 2.9124 2.9729 4.9992 Innovation expenditure 1.0286 1.7998 3.1009 4.4538 4.9938 Results of innovation 1.0007 1.8012 3.0266 4.7134 4.9993 R&D collaboration 1.0005 1.8991 3.1578 4.5982 4.9993 Information sources 1.0459 1.8845 2.9746 4.346 4.9992 Intellectual property rights protection 1.0119 1.9162 2.887 4.2468 4.9992 Purchase and selling of technologies 1.0187 1.9536 3.0489 4.6313 4.9993 Organisational and marketing innovation 1.0033 2.0015 3.2044 4.3163 4.9993 Ecological innovation 1.0311 2.0948 3.1229 4.7682 4.9993 (average score within the cluster; 1 - perfect .. 5 - poor) Diversity reduced to a one-dimensional scale of applicability: Perfect; Good; Average; Average for core concepts and poor for extended framework; Poor
- 8. Perception of the Innovation Survey Concepts (2) 8 Portfolio of the quality of data provided: precise and verified vs. general estimates Perfect accuracy Good accuracy Average accuracy Poor accuracy Cluster Size 0.1935 0.2537 0.309 0.2438 Indicators General firm characteristics (markets, human capital, etc.) 1.0004 1.6399 2.5739 3.8244 Innovation types: product, process, organisational, marketing 1.0226 1.8311 2.8367 5.203 Innovation sales 1.0278 1.8207 3.1647 5.2993 Factors hampering innovation 1.021 2.2062 3.4413 4.6051 Innovation expenditure 1.0005 1.9615 2.7125 5.7329 Results of innovation 1.0923 1.9613 2.9311 5.6131 R&D collaboration 1.0005 1.7636 2.8746 5.8036 Information sources 1.0006 1.8688 3.5205 5.3329 Intellectual property rights protection 1.0006 1.9527 3.3479 5.6574 Purchase and selling of technologies 1.0086 1.9161 3.3584 5.7767 Organisational and marketing innovation 1.0029 2.0603 3.68 5.6648 Ecological innovation 1.0006 2.0853 3.7091 5.7307 (average score within the cluster; 1 - perfect .. 5 - poor) Diversity reduced to a one-dimensional scale: Perfectly precise .. Rough estimates
- 9. Perception of the Innovation Survey Concepts (3) 9 Dimension reduction: joint principle component analysis of applicability and accuracy (rotated component matrix) – 3 dimensions of diversity Core innovation questions Extended questions Accuracy General firm characteristics (markets, human capital, etc.) .822 .257 .125 Innovation types: product, process, organisational, marketing .688 .504 .251 Innovation sales .638 .609 .179 Factors hampering innovation .687 .598 .132 Innovation expenditure .254 .818 .333 Results of innovation .271 .817 .372 R&D collaboration .224 .825 .371 Information sources .306 .816 .253 Intellectual property rights protection .239 .818 .299 Purchase and selling of technologies .244 .815 .393 Organisational and marketing innovation .222 .816 .368 Ecological innovation .180 .812 .417 General firm characteristics (markets, human capital, etc.) .162 .162 .622 Innovation types: product, process, organisational, marketing .112 .254 .851 Innovation sales .245 .298 .800 Factors hampering innovation .602 .216 .636 Innovation expenditure .037 .276 .885 Results of innovation .021 .260 .886 R&D collaboration .016 .254 .904 Information sources .274 .296 .830 Intellectual property rights protection .221 .370 .822 Purchase and selling of technologies .209 .368 .842 Organisational and marketing innovation .217 .344 .833 Ecological innovation .213 .357 .832 Applicabilityandrelevanceof concepts Accuracyofthedataprovided (perfectlyverifiedvs.general estimation) Component • Understanding of core concepts (definitions of innovation and innovation sales) • Understanding the extended framework (ability to account for expenditure on innovation activities, etc. • Quality of data provided
- 10. Portfolio of questionnaire comprehension and data precision 10 Perfect applicability and perfect accuracy Good applicability and accuracy Average applicability, average accuracy Good applicability, poor accuracy Poor applicability, poor accuracy Cluster Size 0.1283 0.2442 0.3473 0.0813 0.1988 Applicability General firm characteristics (markets, human capital, etc.) 1.008 1.6853 2.7245 2.3038 3.1921 Innovation types: product, process, organisational, marketing 1.0006 1.9197 3.0288 2.7295 4.0732 Innovation sales 1.0107 2.026 3.1641 2.6653 4.0347 Factors hampering innovation 1.0156 1.8933 3.0541 1.9834 3.991 Innovation expenditure 1.0006 1.9861 3.0593 2.5826 4.8272 Results of innovation 1.0006 1.9207 3.0286 2.5429 4.8938 R&D collaboration 1.0006 2.0122 3.1561 2.3815 4.9112 Information sources 1.0006 1.981 3.1801 1.7698 4.6812 Intellectual property rights protection 1.0006 1.9325 3.0562 1.8357 4.8577 Purchase and selling of technologies 1.0006 1.97 3.1171 2.3354 4.955 Organisational and marketing innovation 1.004 2.097 3.1603 2.7315 4.7066 Ecological innovation 1.0006 2.1339 3.1202 2.8391 4.9548 Accuracy General firm characteristics (markets, human capital, etc.) 1.0004 1.4989 2.47 3.0271 3.7262 Innovation types: product, process, organisational, marketing 1.0006 1.704 2.6064 4.9991 4.7767 Innovation sales 1.0249 1.6706 2.6115 5.2474 5.309 Factors hampering innovation 1.0289 1.8454 2.9795 4.5602 4.7728 Innovation expenditure 1.0006 1.6296 2.7644 5.4212 5.0025 Results of innovation 1.0007 1.9079 2.7896 5.1648 4.9992 R&D collaboration 1.0006 1.6172 2.7383 5.6144 5.0683 Information sources 1.0092 1.7682 2.8555 4.9991 5.484 Intellectual property rights protection 1.0007 1.6838 2.7808 5.2712 5.8546 Purchase and selling of technologies 1.0007 1.6015 2.8021 5.751 5.8462 Organisational and marketing innovation 1.0042 1.7264 3.1094 5.6506 5.7324 Ecological innovation 1.0007 1.6738 3.1852 5.6957 5.8407 (average score within the cluster; 1 - perfect .. 5 - poor)
- 11. Determinants of questionnaire comprehension 11 Marginal effects of the variables i Perfect applicability and perfect accuracy Good applicability and accuracy Average applicability , average accuracy Good applicabilit y, poor accuracy Poor applicabilit y, poor accuracy Number of employees (log) 0.0370** 0.0234* -0.0297 -0.00267 -0.0281 (0.0162) (0.0121) (0.0279) (0.0237) (0.0176) New to market product innovation 0.1304** 0.0388** 0.0566 0.0149 -0.0631 (0.0506) (0.0161) (0.0312) (0.0057) (0.0858) New to firm product innovation -0.00416 0.0364 -0.0833 0.00228 0.0488 (0.0564) (0.0527) (0.0676) (0.0466) (0.0693) Process innovation -0.0946 0.0625 0.0679 -0.0592 0.0233 (0.0598) (0.0626) (0.0713) (0.0439) (0.0623) Organisational innovation 0.108** -0.103* -0.0907* -0.0432* -0.128** (0.0532) (0.0605) (0.0526) (0.0208) (0.0616) New marketing methods 0.0490 0.0986* -0.158* -0.00809 0.0182 (0.0533) (0.0554) (0.0818) (0.0554) (0.0646) Ongoing innovation -0.123* -0.00512 0.00673 0.0470 0.0740 (0.0699) (0.0665) (0.0769) (0.0472) (0.0677) Abandoned innovation -1.064*** -0.0398 0.842*** 0.189** 0.0722 (0.199) (0.174) (0.168) (0.0951) (0.151) Technologies for automation and logistics -0.00160 -0.0291 -0.0345 0.0354 0.0298 (0.0230) (0.0305) (0.0373) (0.0338) (0.0359) Advanced production processes 0.0806** 0.0469* 0.0225 0.0313 -0.101 (0.0238) (0.0246) (0.0389) (0.0267) (0.0821) Optimized organisational concepts 0.000690 0.0874*** -0.0278 -0.00330 -0.0570 (0.0304) (0.0279) (0.0442) (0.0250) (0.0384) Innovation management 0.0658*** 0.0321** 0.0370** -0.0295 -0.105*** (0.0224) (0.0164) (0.0125) (0.0274) (0.0291) Has in-house innovation effort 1.191*** 0.629*** -0.191 -0.186* -0.185* (0.206) (0.127) (0.168) (0.107) (0.0753) Active at the international market 0.0213** 0.0805 0.0934 -0.0594 -0.136** (0.0079) (0.0621) (0.0654) (0.0453) (0.0644) Base questionnaire fill-in strategy: only accounting dept Economic and financial planning depts 0.141* 0.0402 -0.235*** 0.0585 -0.00492 (0.0743) (0.0664) (0.0878) (0.0704) (0.0763) Top management and technical specialists 0.0898 0.0890 -0.112 0.00837 -0.0753 (0.0588) (0.103) (0.0935) (0.0686) (0.0581) Complex: economic planning, technical depts and others -0.0507 0.0873 -0.348*** -0.0331 0.344*** (0.0355) (0.0790) (0.0751) (0.0499) (0.0590) All types of departments -0.0403 -0.116* -0.464*** -0.0559 0.677*** (0.0508) (0.0633) (0.0641) (0.0394) (0.0552) Number of employees to complete the survey -0.0747* 0.0213* 0.289*** 0.00380 -0.240*** (0.0443) (0.0158) (0.0491) (0.0355) (0.0544) Observations 401 401 401 401 401 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Marginal effects after mlogit choice regression: comprehension of the innovation survey questionnaire Firm size: larger companies Innovation: new-to-market innovation but even more important: organisational innovation Technological level: Advanced production technologies but even more important: Innovation management culture In-house innovation effort Activity at the international market Companies better at providing data: Best organisation of data collection: • Economic and financial planning depts. • Limited number of responsible employees
- 12. Conclusions 12 (1) Grading the comprehension of the innovation survey and the accuracy of data provision: • understanding basic definitions • ability to measure broader indicators • accuracy of data provided (2) Advanced companies are better at collecting and delivering innovation-related data • New-to-market product innovators • International markets • Advanced technology levels (3) But organizational and innovation culture is of higher importance • Organisational innovation • Advanced organisational concepts • Innovation management culture Key findings (1) Need for clear guidelines (or sections) • Explaining basic definitions • Recommendations on identifying complex indicators • Instructions to increase accuracy (2) Overcome existing bias towards advanced and large companies • Modular surveys should be more friendly enterprises notable for to less sophisticated innovation strategies and ad-hoc innovation management processes (3) Survey guidelines tailored for different enterprise units • accountants • technical specialists • top management Implications