Title
Community Innovation Survey 4
The Community Innovation Survey (CIS) is a survey of innovation activity in enterprises covering EU Member States, candidate countries, Iceland and Norway.
In order to ensure comparability across countries, Eurostat, in close cooperation with the EU Member States and other countries, developed standard core questionnaires for CIS 4, with an accompanying set of definitions and methodological recommendations.
CIS 4 is based on the Oslo Manual (2nd edition, 1997), which gives methodological guidelines and defines the concept of innovation, and on Commission Regulation No 1450/2004.
Geographical Coverage
16 countries: Belgium, Bulgaria, Czech Rep., Germany, Estonia, Spain, Greece, Hungary, Italy, Lithuania, Latvia, Norway, Portugal, Romania, Slovenia, Slovakia
Time Period Covered
2002-2004, reference year is 2004
-
Country Specific Information: CIS 4
-
CZ - Czech Republic
Corresponding National Study
Technical Innovation in the Czech Republic
Producer
Czech Statistical Office
Universe
Units with more than 10 employees and from the selected manufacturing and service (financial or not financial) sectors, that according to their prevailing activity belong to the NACE rev.1:
0-14 Mining and quarrying
Manufacturing (15-37)
15-16 Food, beverages and tobacco
17-19 Textile and leather
20-22 Wood, pulp and printing
23-24 Coke and chemicals
25-26 Rubber and other non-metallic
27-28 Basic and fabricated metals
29 Machinery and equipment
30-33 Electrical and optical equipment
34-35 Transport equipment NEC
36-37 Manufacturing NEC and recycling
40-41 Electricity, gas and water supply
Services (51,60-67,72-73,74.2,74.3)
51 Wholesale trade
60-64 Transport, storage and communication
65-67 Financial intermediation
72 Computer and related activities
73 Research and development
74.2 Architectural and engineering activities
74.3 Technical testing and analysis
Source Of Sampling Frame
Czech Business Register
Sampling Design
combination of stratified sampling and census
Sampling Method
metadata not available
Units Of Observation
- Enterprise
Date of Data Collection
2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
metadata not available
Resources
-
Technical Innovations in the Czech Republic – 2002-2003. Available at link.
-
LV - Latvia
Corresponding National Study
Research and Innovation
Producer
Central Statistical Bureau of Latvia
Universe
Enterprises from the domains:
- mining and quarrying (NACE 10-14)
- manufacturing (NACE 15-37)
- electricity, gas and water supply (NACE 40-41)
- wholesale trade (NACE 51)
- transport, storage and communication (NACE 60-64)
- financial intermediation (NACE 65-67)
- computer and related activities (NACE 72)
- architectural and engineering activities (NACE 74.2)
- technical testing and analysis (NACE 74.3)The target population includes also NACE 01,02,05, 45 -50, 52-55, 70-71, 73, 74.1, 74.4-74.8, 75-93.
Latvia included enterprises with 0 - 9 employees. Employees 0 and 1-9 only for NACE 73 and
NACE 7413.or activity-in-innovation identification (research projects done in 2003 or innovative activity in entrepreneurship in 2003 or 2004: on Yes/No basis).
Source Of Sampling Frame
The sample was drawn from the frame population that was up-to-date in November 2004. The population frame of survey
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
The sample design is a stratified simple random sample. Enterprises were stratified using four indicators: number of employees (8 size classes); NACE group (83 groups); region (6 regions); activity-in -innovation identification (research projects done in 2003 or innovative activity in entrepreneurship in 2003 or 2004: on Yes/No basis). Number of strata is 1599. Number of samples is 1.
All big statistical units (employees >249 in size group 5) are included in the sample. Statistical units with employees 0 (size group 6) and 1-9 (size group 7) only for NACE 73 and NACE 7413 are all included in the sample or activity-in-innovation identification (research projects done in 2003 or innovative activity in entrepreneurship in 2003 or 2004: on Yes/No basis).
Other statistical units are included by choice (simple random sample).
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
Used the inverse of the sampling fraction as the weights Nh/nh, where Nh is the population in stratum h and nh is the number of enterprises in the realised sample in stratum h, assuming that each unit in the stratum had the same inclusion probability. This will automatically adjust the sample weights of the respondents to compensate for unit non-response. If the size group is 5 (number of employees 249+), then always weights = 1 in the sample and in the realised sample.
Software used:
SUDAAN 7.5.2, SPSS 11.5.1, EXCEL 2000.
Resources
-
-
LT - Lithuania
Corresponding National Study
The 4th Community Innovation Survey
Producer
Statistics Lithuania
Universe
All enterprises with 10 employees or more, from:
- mining and quarrying (NACE 10-14)
- manufacturing (NACE 15-37)
- electricity, gas and water supply (NACE 40-41)
- wholesale trade (NACE 51)
- transport, storage and communication (NACE 60-64)
- financial intermediation (NACE 65-67)
- computer and related activities (NACE 72)
- architectural and engineering activities (NACE 74.2)
- technical testing and analysis (NACE 74.3)No deviations from the
section 2 of the annex of the Commission Regulation No 1450/2004.
Source Of Sampling Frame
The official, up-to-date, statistical business register of the country.
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
The sample were divided into strata according NACE at 2 digit level (except for NACE 74: The 3 digit sections NACE 74.2 and 74.3 were treated as separate NACE categories while NACE 74.1 and 74.4 to 74.8 were treated as a single NACE category.) and enterprise size (small, medium, large).
There was used stratified random sampling. The final number of strata is 122.
The Neyman optimal allocation was used for sample selection and determination of the sample size.
If stratum size is 16 or less then all enterprises in this stratum are included.
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
The non-respondent units were assumed to resemble those who have responded to the survey and treated as non-selected units. For this, the weighting or the grossing up factors were adjusted: the design weight Nh / nh is replaced by Nh / mh where Nh is the size of stratum h, nh is the sample size in stratum h and mh is the number of respondents in stratum h.
The software package CLAN was used for calculations.
Resources
-
CIS 4 The 4th Community Innovation Survey. Quality Report for Country Lithuania (version 14 August 2006).
-
SI - Slovenia
Corresponding National Study
4th Community Innovation Survey
Producer
Statistical Office of the Republic of Slovenia
Universe
All enterprises with 5 employees or more, from the industries mentioned in the
section 2 of the annex of the Commission Regulation No 1450/2004.
In addition, there were additional "non-core" industries:
- research and development (NACE 73)
- construction (NACE 45)
- motor trade (NACE 50)
- retail trade ( NACE 52)
- legal, acccounting,market research,consultancy, management services (NACE 74.1)
- advertising (NACE 74.4)
- labour recruitment and provision of personnel (74.5)
- investigation and security activities (NACE 74.6)
- industrial cleaning services (NACE 74.7)
- miscelaneous business activities n.e.c. (NACE 74.8)
- real estate activities (NACE 70)
- hotels and restaurants (NACE 55)
- renting of machinery and equipment without an operator 8NACE 71)
Source Of Sampling Frame
The official, up-to-date, statistical business register was used.
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
There were 173 strata in sampling frame according to the 2 digit NACE (and to the 3 digit NACE 74.2 and 74.3) and 4 size classes of number of employees. All units with 50 or more employees were selected with certainty. Among other units approximately 14% units from each strata were selected taking into account that in each strata were selected at least min{number of units in strata, 6} units.
Data for the CIS 4 survey were collected through combination of census and a sample survey; in the sample small enterprises from 5-49 employees were included. Other enterprises (medium-sized and large) were all included in the survey (census).
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
Simulations were performed in order to see if there is some difference between respondents and non-respondents, but we found out that there is no significant difference between them. So we employed standard formulas for calculation of non-response weights in each stratum.
Weights were calculated by using module SAS-base.
Resources
-
CIS 4 The 4th Community Innovation Survey. Quality Report for Country Slovenia (version 25 August 2006).
-
SK - Slovakia
Corresponding National Study
Statistical survey on innovations (INOV) 2002-2004
Producer
Statistical Office of the Slovak Republic
Universe
metadata not available
Source Of Sampling Frame
metadata not available
Sampling Method
metadata not available
Date of Data Collection
2002-2004
Weighting: Method
metadata not available
Resources
-
-
RO - Romania
Corresponding National Study
Innovation in industry and services during 2002-2004
Producer
National Institute of Statistics
Universe
All enterprises with 10 employees or more, from:
- mining and quarrying (NACE 10-14)
- manufacturing (NACE 15-37)
- electricity, gas and water supply (NACE 40-41)
- wholesale trade (NACE 51)
- transport, storage and communication (NACE 60-64)
- financial intermediation (NACE 65-67)
- computer and related activities (NACE 72)
- architectural and engineering activities (NACE 74.2)
- technical testing and analysis (NACE 74.3)Additionally, the NACE 73 (research and development) was introduced.
Source Of Sampling Frame
The official, up-to-date, statistical business register of the country.
Sampling Design
stratified sampling
- random sampling within strata
Sampling Method
The sampling design used was the stratified sampling with simple random sampling within the strata. The strata were defined according to the geographical region, economical activity and the size of the enterprise. For the sample allocation we used the Neymann allocation method
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
The calibration method was CLAN.
Resources
-
CIS 4 The 4th Community Innovation Survey. Quality Report for Country Romania.
Available at this link.
-
IT - Italy
Corresponding National Study
Community Innovation Survey (CIS4)
Producer
Italian National Institute of Statistics (ISTAT)
Universe
All enterprises included in the target population follow the minimum coverage which is all enterprises with 10 employees or more, from the following industries are included in the core target population of the CIS4:
- mining and quarrying (NACE 10-14);
- manufacturing (NACE 15-37);
- electricity, gas and water supply (NACE 40-41);
- wholesale trade (NACE 51);
- transport, storage and communication (NACE 60-64);
- financial intermediation (NACE 65-67);
- computer and related activities (NACE 72);
- architectural and engineering activities (NACE 74.2);
- technical testing and analysis (NACE 74.3).
"Non-core" industries that were covered in addition are:
- construction (NACE 45);
- motor trade (NACE 50);
- retail trade (NACE 52);
- hotels and restaurants (NACE 55);
- real estate activities (NACE 70);
- renting of machinery and equipment without an operator (NACE 71);
- research and development (NACE 73);
- other business activities: legal, accounting, book-keeping and auditing activities; tax consultancy; market research and public opinion polling; business and management consultancy; holdings (NACE 74.1); advertising (NACE 74.4); labour recruitment and provision of personnel (NACE 74.5); investigation and security activities (NACE 74.6); industrial cleaning (NACE 74.7); miscellaneous business activities n.e.c. (NACE 74.8).
Source Of Sampling Frame
The official, up-to-date, statistical business register, called ASIA (Archivio Statistico delle Imprese Attive - statistical business register of active enterprises) was used.
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
The target population of the CIS4 is broken down into similar structured subgroups or strata (which should be as homogeneous as possible and form mutually exclusive groups). The stratification variables to be used for the CIS4, i.e. the characteristics used to break down the sample into similarly structured groups, are:
1. the economic activities (in accordance with NACE). Stratification by NACE has been done at least at two-digit (division) level, except for NACE 24, 35, 74. NACE 24.4, 35.3, 74.2 and 74.3 groups were treated as separate NACE sectors while the remaining groups of NACE 24, 35 and 74 were treated as single NACE sectors;
2. enterprise size according to the number of employees. The size-classes used were the following: 10-49 employees; 50-249 employees; 250+ employees.;
3. regional aspects at NUTS 2 level. The regional allocation of the sample was taken into consideration when sampling. In particular, the breakdown of national territory into regions was performed on the basis of the NUTS level 2. (Regulation EC No 1059/2003 of the European Parliament and of the Council of 26 May 2003 on the establishment of a common classification of territorial units for statistics - NUTS)
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
Calibration estimators methodology, currently applied at Istat, was used for the estimation process. It can be applied to the extent that the known totals of some auxiliary variables, strictly correlated to the variables of interest, are available. These calibration estimators have the following properties: they are more efficient than the direct estimators because of the auxiliary information used; they reduce the bias effect due to the non response and under coverage; they produce estimates of auxiliary variables that equal the known totals of such variables.
The final weights are obtained by adopting the following procedure: an initial weight is assigned to each sampled unit with reference to the sampling plan as the reciprocal of the inclusion probability. Two correction factors for initial weights are then calculated: a first one is the unit non response factor; a second one is to satisfy equality between estimation
of auxiliary variables and known totals from the Register. The final weights are thus obtained as the result of the product between initial weights and correction factors.
For CIS, as well as for most of the business surveys, number of enterprises and number of employees were used as auxiliary variables, according to the information provided by the Italian Official Business Register ASIA.
The results of the non-response analysis have also been taken into account when calculating the final weighting factors applying the methodology adopted by Hansen e Hurwitz in order to reduce the non-response bias of the estimator concerning the number of innovating firms.
Resources
-
Community Innovation Survey (CIS4). Microdata File for Research, Survey Methodology (November 2008).
Available at link.
-
NO - Norway
Corresponding National Study
Fourth Community Innovation Survey (CIS 4)
Producer
Statistics Norway
Universe
All enterprises with 10 employees or more, from:
- mining and quarrying (NACE 10-14)
- manufacturing (NACE 15-37)
- electricity, gas and water supply (NACE 40-41)
- wholesale trade (NACE 51)
- transport, storage and communication (NACE 60-64)
- financial intermediation (NACE 65-67)
- computer and related activities (NACE 72)
- architectural and engineering activities (NACE 74.2)
- technical testing and analysis (NACE 74.3)In addition to the core industries, NACE 05.02, 45 and 73 were included.
Source Of Sampling Frame
The official, up-to-date, statistical business register of the country.
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
All units with 50 employees or more were included in the survey*, whereas a sample were drawn for enterprises with 10-49 employees. The samples were based on the size of the different industries. For larger industries, i.e. industries with many enterprises, a smaller percentage of the units were drawn. Specifically the sample rate for enterprises with 10-49 employees was 15 per cent for NACE 15, 45, 51 and 60. For all other NACE/ industries the sample rate for enterprises with 10-49 employees was 35 per cent.
Units Of Observation
- Enterprise
Date of Data Collection
2003-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
- Web survey
Weighting: Method
Statistics Norway used SAS-macro commands developed by its own staff.
Resources
There is an exception to the rule of total coverage for enterprises with 50 employees or more; out of all enterprises within NACE 45 and 51 with 50-99 employees, a sample of 35 per cent were drawn, due to the large number of enterprises in these NACE-classes. Beyond that the sampling was carried out in accordance with sub-chapter 2.4.
CIS 4 The 4th Community Innovation Survey. Quality Report for Norway (version 14 August 2006).
-
HU - Hungary
Corresponding National Study
The 4th Community Innovation Survey
Producer
Hungarian Central Statistical Office
Universe
All enterprises included in the target population follow the minimum coverage which is all enterprises with 10 employees or more from the following industries:
mining and quarrying (NACE 10-14)
manufacturing (NACE 15-37)
electricity, gas and water supply (NACE 40-41)
wholesale trade (NACE 51)
transport, storage and communication (NACE 60-64)
financial intermediation (NACE 65-67)
computer and related activities (NACE 72)
architectural and engineering activities (NACE 74.2)
technical testing and analysis (NACE 74.3)NACE 45 and NACE 73 were covered in addition
Source Of Sampling Frame
The official, up-to-date, statistical business register of the country was used.
Sampling Design
stratified sampling
- random sampling within strata
Sampling Method
A stratified sample was selected. The stratification variables were economic activity (categories: group 24.4 and the remaining groups in division 24; groups 35.1, 35.3 and the remaining groups in division 35; groups 74.2, 74.3 and the remaining groups in division 74; divisions elsewhere) and size (categories: >249, from 150 to 249, from 100 to 149, from 50 to 99, from 20 to 49 and from 10 to 19 employees). Thus the number of strata was 47*6=282.
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
No calibration was made.
Resources
-
CIS 4 The 4th Community Innovation Survey. Quality Report for Hungary (version 25 August 2006)
-
PT - Portugal
Corresponding National Study
CIS 4 - Community Innovation Survey
Producer
Statistics Portugal
Universe
All enterprises with 10 employees or more, from:
- mining and quarrying (NACE 10-14)
- manufacturing (NACE 15-37)
- electricity, gas and water supply (NACE 40-41)
- wholesale trade (NACE 51)
- transport, storage and communication (NACE 60-64)
- financial intermediation (NACE 65-67)
- computer and related activities (NACE 72)
- architectural and engineering activities (NACE 74.2)
- technical testing and analysis (NACE 74.3)Apart from the core industries, Norway surveyed also NACE 45, 52, 55, 71, 73, and 74.1 to 74.8 (except 74.2 and 74.3).
There were some deviations: on NACE 24.2, 30, 32, 33, 35.3, 64, 72, 73, 74.2 and 74.3 Norway covered enterprises with 5 employees or more, on NACE 45 and 52 Norway covered enterprises with 50 employees or more and on NACE 55 Norway covered enterprises with 250 employees or more.
Source Of Sampling Frame
The official, up-to-date, statistical business register of the country.
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
Norway used the three classification variables mentioned before. The sample was prepared by our National Statistics Office accordingly with the methodological guidelines proposed by Eurostat. The number of strata was equal to 639 and the sample was equal to 7.370 enterprises.
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Web survey
Weighting: Method
For the sample weights we used the inverse of the sampling fraction (using the number of enterprises).
No calibration method was used.
Resources
The sampling frame was established by the National Statistical Office according to the criterion defined by Eurostat and the national statistical record of the Portuguese enterprises. However, despite the many efforts made by OCES, some enterprises could not be retraced: they totalized 256, which represented 3,7% of the original sample. They were withdrawn from the sample and from the population and the sample was readjusted.
CIS 4 The 4th Community Innovation Survey. Quality Report for Country Portugal (version 25 August 2006).
-
BG - Bulgaria
Corresponding National Study
The statistical survey on innovation activity
Producer
National Statistical Institute
Universe
metadata not available
Source Of Sampling Frame
metadata not available
Sampling Design
complete census
Sampling Method
metadata not available, but Bulgaria seems to have employed a complete census in the upcoming rounds
Units Of Observation
- Enterprise
Date of Data Collection
2005
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
metadata not available
Resources
-
Eurostat (2007) Community Innovation Statistics - Fourth Community Innovation Survey (CIS 4) and European Innovation Scoreboard (EIS) 2006 - Issue number 116/2007.
/nAvailable at this link.
-
EL - Greece
Corresponding National Study
Study for the measurement of Innovation in Greek Enterprises (2002 – 2004)
Producer
Quantos S.A. and MRB Hellas
Universe
The target population of the survey was the Core target population described in the Annex of Commission Regulation No 1450 / 2004.
Enterprises with less than 10 employees during the period of study were excluded from the target population.
Source Of Sampling Frame
metadata not available
Sampling Design
stratified sampling
- random sampling within strata
Sampling Method
Stratified sampling with optimal allocation was used for the selection of a sample of 2.648 units (enterprises) from the 11.967 units of the target population.
The stratification variables were the enterprises’ economic activity, size and geographical location of their headquarters in Greece.
The stratification with regards to the first two stratification variables was implemented by using the categories described in Eurostat’s text of methodological recommendations, but this could not have been applied in the case of the third stratification variable.
Instead of the geographical stratification of the enterprises into the 13 Nuts II regions, three regions were used for the stratification based on the location of their head office: Attiki, Kentriki Makedonia and Rest of Greece.
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
The methodological recommendations of Eurostat were used for the calculation of the weights. For estimations concerning quantitative variables (i.e. turnover, number of employees, etc.) the ratio of the total turnover in the population and in the released sample for each stratum was used, while for estimations concerning qualitative
variables the number of enterprises was used instead of their turnover.
The weights were recalculated after the evaluation of the results of the non-response analysis in order to adjust for the bias deriving from non – response.
Resources
-
-
ES - Spain
Corresponding National Study
Technological Innovation in Companies Survey
Producer
Instituto Nacional de Estadistica (Spanish Statistical Office)
Universe
All industrial, construction and services enterprises with at least ten employees.
Source Of Sampling Frame
metadata not available
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
metadata not available
Units Of Observation
- Enterprise
Date of Data Collection
2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
metadata not available
Resources
-
-
BE - Belgium
Corresponding National Study
The 4th Community Innovation Survey
Producer
Scientific and Technical Information Service (STIS)
Universe
All enterprises with 10 employees or more, from the following industries:
- mining and quarrying (NACE 10-14)
- manufacturing (NACE 15-37)
- electricity, gas and water supply (NACE 40-41)
- wholesale trade (NACE 51)
- transport, storage and communication (NACE 60-64)
- financial intermediation (NACE 65-67)
- computer and related activities (NACE 72)
- architectural and engineering activities (NACE 74.2)
- technical testing and analysis (NACE 74.3)
Source Of Sampling Frame
metadata not available
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
Data were collected through both a mail survey and an electronic online questionnaire(firmswere sent a paper booklet by regular mail and could opt to respond either on paper or online). A mixture of census and sample survey was used, according to the stratum under consideration:
- For the Brussels Region, a census was performed for allenterprises in the core industries (NACE 10-41, 50,60-72, 74.2 and 74.3) and in NACE 50. Large (250 or more employees) and medium-sized (between 50 and 249 employees) firms in the non-core industries 45, 52 and 74 (exc. 74.2 and 74.3) were also exhaustively sampled, whereas small enterprises (between 10 and 49 workers) in each of these industries were sampled at random;
- For the Walloon Region, a census was performed for all medium-sized (between 50 and 249 employees) and large (250 or more employees) enterprises. Small enterprises (between 10 and 49 workers) were sampled at random within each industry;
- For the Flanders Region, census sampling was done for large size firms (i.e. with 250 or more employees), and stratified random sampling was used for the other enterprises.
Units Of Observation
- Enterprise
Start Date
01-2002
End Date
12-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
- Web survey
Weighting: Method
metadata not available
Resources
-
Eurostat (2007) Community Innovation Statistics - Fourth Community Innovation Survey (CIS 4) and European Innovation Scoreboard (EIS) 2006 - Issue number 116/2007.
-
DE - Germany
Corresponding National Study
4th Community Innovation Survey
Producer
Centre for European Economic Research - ZEW
Universe
All enterprises with 10 employees or more, from the following core industries:
- miningandquarrying(NACE10-14)
- manufacturing(NACE15-37)
- electricity,gasandwatersupply(NACE40-41)
- wholesaletrade(NACE51)
- transport,storageandcommunication(NACE60-64)
- financialintermediation(NACE65-67)
- computerandrelatedactivities(NACE72)
- architecturalandengineeringactivities(NACE74.2)
- technicaltestingandanalysis(NACE74.3)
Source Of Sampling Frame
The database of a credit rating agency (CREDITREFORM), which contains information on virtually all enterprises with more than 5 employees in Germany enterprises.
Sampling Design
combination of stratified sampling and census
- random sampling within strata
Sampling Method
The size classes used for stratification are: 5-9, 10-19, 20-49, 50-99, 100-249, 250-499, 500- 999, 1000+
Generally, the industry classification used for stratification is the two-digit NACE level.
We split up NACE 22 in the two categories 22.1 (publishing) and 22.2 + 22.3 (printing), NACE 24 in the two categories 24 without 24.4 (chemistry without pharmaceuticals) and 24.4 (pharmaceuticals) and NACE 64 in the two categories 64.1 (post) and 64.2 (telecommunication).
In addition to being stratified by NACE and size, our sample will also be stratified by East and West Germany.
We use random sampling techniques with known and disproportional selection probabilities, i.e. larger enterprises and enterprises in East Germany have a higher selection probability.
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
A non-response analysis was carried out to calculate unit non-response corrected weighting factors according to the methodological recommendations from Eurostat. Two different weights are used for qualitative and quantitative variables:
Weights based on enterprises: all qualitative variables
Weight based on number of employees or weight based on turnover: quantitative variables.
Resources
-
SEW (2006) The 4th Community Innovation Survey. Quality Report for Germany
-
EE - Estonia
Corresponding National Study
Innovation in Estonian Enterprises
Producer
Statistics Estonia
Universe
The survey sample included enterprises of mining industry, manufacturing industry and the services sphere. It should be stated in advance that the results mainly reflect the latter two, while the mining industry has been reviewed only superficially. It is also important to note that regional analysis has not concentrated on details, since the spheres of activity need not be statistically representative in regional breakdown.
Source Of Sampling Frame
metadata not available
Sampling Method
metadata not available
Units Of Observation
- Enterprise
Date of Data Collection
2002-2004
Type Of Data Source
Survey
Interview Mode
- Postal survey
Weighting: Method
metadata not available
Resources
-
Terk, E. et all (2007). Innovation in Estonian Enterprises. Based on the Estonian results of the Fourth Community Innovation Survey (CIS 4). – Innovation Studies, No 7. Tallinn: Enterprise Estonia.