See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/345499344

Sources of innovation and innovation type: firm-level evidence from the United States
Article in Industrial and Corporate Change · December 2019
DOI: 10.1093/icc/dtz010

CITATIONS READS

35 1,088

3 authors:

Mehmet Akif Demircioglu David B. Audretsch

National University of Singapore Indiana University Bloomington
49 PUBLICATIONS 952 CITATIONS 760 PUBLICATIONS 69,853 CITATIONS

SEE PROFIL E SEE PROFILE

Tim Slape r
Indiana University Bloomington
17 PUBLICATIONS 1,128 CITATIONS

SEE PROFIL E

Some of the authors of this publication are also working on these related projects:

social media View project

The university technology transfer revolution in Saudi Arabia View project

All content following this page was uploaded by Mehmet Akif Demircioglu on 05 January 2021.

The user has requested enhancement of the downloaded file.

 Industrial and Corporate Change, 2019, Vol. 28, No. 6, 1365–1379
doi: 10.1093/icc/dtz010
Advance Access Publication Date: 15 March 2019
Original article

Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November 2020

Sources of innovation and innovation type: firm-level
evidence from the United States
Mehmet Akif Demircioglu ,1,* David B. Audretsch2 and Timothy F. Slaper3
1
Lee Kuan Yew School of Public Policy, National University of Singapore, 469B Bukit Timah Road, Level 2
Li ka Shing Building, Singapore 269771, Singapore. e-mail: mehmet@nus.edu.sg, 2School of Public and
Environmental Affairs, Indiana University, Bloomington, IN 47405, USA. e-mail: daudretsch@indiana.edu
and 3Bloomington Kelley School of Business, Indiana Business Research Center, Indiana University,
Bloomington, IN 47405, USA. e-mail: tslaper@indiana.edu *Main author for correspondence.

Abstract
Only a handful of studies on innovation empirically analyze the links between firm innovation and the sources of
that innovative activity of sources of innovation on types of innovation. To fill this gap in the literature, this study
provides one of the first tests to identify how important sources of new information (suppliers, customers, other
business people in the industry, workers, and university) are associated with types of innovations (product, process,
and marketing). Data come from the 2014 National Survey of Business Competitiveness sponsored by the Economic
Research Service at the United States Department of Agriculture (n¼10,952). The results show that innovation ideas
emanating from customers, workers, and universities are positively associated with all types of innovations,
suggesting that these sources are critical for developing different types of innovation. In particular, universities as a
source of innovation activity are especially important. In contrast, other sources, such as suppliers and people in
industry do not seem to be as important as a source of innovation. JEL classification: C18, C21, L10, O10, O30, O31,
O32

1. Introduction
Innovation is one of the most discussed topics in the fields of business, economics, and management studies. Innovation can
help an organization to adapt to the environment, survive, and grow (Burns and Stalker, 1961; Thompson, 2010), enhance
economic growth and performance (Carlino, 2001; Verspagen, 2006), narrow the gap in productivity and economic
conditions for less successful countries (Fagerberg and Godinho, 2006), increase competitiveness (Cantwell, 2006), raising
living standards and prosperity of citizens (Aboal and Tacsir, 2018; Breznitz et al., 2018: 883), help to reduce unemployment
(Pianta, 2006), and help to deal with grand challenges such as poverty, climate change, and health (Kattel and Mazzucato,
2018; Mazzucato, 2018). Overall, innovation is a driving force of firm productivity (Stojcic and Hashi, 2014) and “central to
the growth of output and productivity” of firms and nations (OECD/Eurostat, 2005: 10; see also –>Leyden and Link, 1992;
Link and Siegel, 2007; Acs et al., 2017).

VC The Author(s) 2019. Published by Oxford University Press on behalf of Associazione ICC. All rights reserved.
1366 M. A. Demircioglu et al.
Studies of innovation are roughly divided into two types. First, many studies treat innovation as a dependent variable, so
they have examined factors affecting innovation such as determinants or conditions of innovations (e.g. Mohr, 1969;
Kimberly and Evanisko, 1981; Damanpour, 1991; Greenhalgh et al., 2004; Torugsa and Arundel, 2016a; Demircioglu and
Audretsch, 2017a). The second type of study considers innovation as an independent variable, and examines the outcomes of
innovation, such as how innovation influences efficiency, increases performance, improves the quality of services, and even

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
increases the legitimacy of organizations (Damanpour et al., 2009; Griffith et al., 2006; Verhoest et al., 2007; Brown and
Osborne, 2012; Ballot et al., 2015). This article uses the first approach and treats innovation as the dependent variable while
analyzing how different sources associated with overall firm innovation activity and four types of innovations.
The magnitude and breadth of the literature on innovation and particularly sources of innovation reflects its importance.
The famous conclusion attributed to Solow that innovation “falls like manna from heaven” ( Aghion and Howitt, 1998;
Link, 2013), may in fact ignore that innovation activity emanates from more earthly sources. Analyzing two types of
knowledge spillovers—Marshall-Arrow-Romer (MAR) and Carlino (2001) point out that analyzing different sources of
innovation goes back to 1890 when Alfred Marshall developed a theory on knowledge and innovation externalities. For
instance, Nakamura (2000) explains how workers have been a key source of creativity and innovations for the last century.
Additionally, research facilities and universities provide a source of knowledge and ideas underlying the development of
many product and service innovations (Carlino, 2001). Thus, innovation sources are a crucial factor for innovation activities
and different sources of innovation may result in different types of innovations.
Despite its central importance, there is only a paucity of studies actually attempting to uncover what exactly are the most
important sources of innovation as well as their impacts on overall innovation activity and the various dimensions and
manifestations of innovation. The handful of empirical studies found in the literature are typically restricted to descriptive,
conceptual, qualitative analysis, or based on case studies (e.g. Borins, 2002; OECD, 2009; Birkland, 2011; Nasi et al.,
2015; Bankins et al., 2017). This paucity of quantitative studies identifying and analyzing sources of new information and
knowledge driving innovation activity limits both the ability of scholars to not only understand the innovation process but
also thought leaders to formulate effective policies and strategies in policy and business. The purpose of this article is to
make a least a start toward filling this gaping gap in the literature by analyzing how particular sources of knowledge impact
both overall innovation activity as well as specific manifestations of innovation output. In particular, this article seeks to
analyze how the important sources of knowledge, such as suppliers, customers, other people in industry, workers, and
universities, are associated with specific types of innovations, namely, product, process, and marketing innovations. To
address these questions, we analyze which particular knowledge sources generate particular types of innovations and
overall innovation activity using quantitative methods.
The second important contribution of this study is to compare three distinct types of innovation—product, process, and
marketing innovations, instead of just focusing on a grossly simplistic conceptual and aggregate measurement of innovation
(e.g. whether innovation occurs or not). Despite difficulties in quantifying and measuring innovation activity, it is possible
to measure different types of innovations (OECD/Eurostat, 2005, 2018; Smith, 2006; Bloch and Bugge, 2013; Bugge and
Bloch, 2016; Demircioglu and Audretsch, 2018; Gault, 2018). Most of the extant research typically measures aggregated
innovation activity by patents, as an outcome variable (e.g. Arundel and Kabla, 1998; Arundel, 2001). However, patents as
an output does not capture many innovations. For instance, Arundel and Kabla (1998: 138) state that “Patents are a
particularly poor measure of innovativeness in sectors such as food and tobacco, petroleum refining, basic metals,
automobiles, and other transport equipment. In these sectors, the large majority of innovations are not patented.” In fact,
many innovations which are not patented (or aimed to be patented) are measured using the large dataset introduced in this
article and explained below.
Additionally, three specific types of innovation exhibit different characteristics from each other and may emanate from
distinct sources of knowledge. For example the knowledge sources underlying marketing innovations may be different than
their counterparts for process innovations [see statistical results from the Policies, Appropriation, and Competitiveness in
Europe (PACE), Service des Statistiques Industrielles (Statistical Service of the French Ministry of Industry; SESSI),
Community Innovation Surveys (CIS), and Terjesen and Patel, 2017]. Thus, it makes sense to analyze these three types of
innovations differently. For instance, the 1993 CIS survey showed that process innovation was about 25%, which was lower
than product innovation (35.9%; Arundel and Kabla, 1998).
The third important contribution to the literature from this article is the unique and representative dataset for the US data
that are analyzed. Despite the high propensity for innovation activity to spatially cluster within close geographic proximity to
the knowledge source, resulting in notable innovative clusters, such as Silicon Valley, no systematic, comprehensive, and
large-scale survey on innovation activity had been created for the United States spanning the entire economy. In contrast,
Sources of innovation and innovation type 1367
Europe has devoted considerable resources and effort in creating and analyzing systematic and comprehensive measurement
of innovation activity. Most prominently are the Oslo Manuals, the European Union’s Community Innovation Surveys (CIS)
conducted in different European countries, the Maastricht Economic and Social Research Institute on Innovation and
Technology (MERIT), MEPIN, and SESSI, resulting in a number of important and influential scholarly and policy
publications (e.g. Arundel and Kabla, 1998; Arundel, 2001; Arundel and Geuna, 2004; OECD/ Eurostat, 2005, 2018; Frenz
and Ietto-Gillies, 2009; Battisti and Stoneman, 2010; Bloch and Bugge, 2013).

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
The limited datasets measuring innovation in the United States (e.g. National Science Foundation’s Survey of Industrial
Research and Development, CompuStat, and 1994 Carnegie Mellon University survey of manufacturing innovation) suffer
in that they are neither comprehensive nor representative of American companies, and do not reflect the diverse and
heterogeneous nature of both knowledge sources or types of innovations (Cooper and Merrill, 1997). Data limitations have
restricted the ability of scholars to undertake comprehensive and wide-ranging studies on innovation in the context of the
United States. Thus, an additional key contribution of this study is to provide one of the first comprehensive, representative,
and systematic analyses of innovation in the context of the United States by using the National Survey of Business
Competitiveness (NSBC) data.
In particular, the novel contribution of this article to the literature is to compare three different types of innovations
(product, process, and marketing) to show how five important knowledge sources (suppliers, customers, other people in
industry, workers, and universities) are associated with these three innovation types as well as overall firm innovation
activity differently. The main research question of this article thus is how do important sources of knowledge are associated
with main innovation types and overall innovation activity? There are two sub-research questions in this study. First, how
do suppliers, customers, and other business people in the industry, workers, and university as sources of innovation
contribute to product, process, and marketing innovations? Second, how do these knowledge sources contribute to the
overall firm innovation activity (the aggregate of all types of innovations)? This article is organized as follows. The
following section will explain and discuss sources and types of innovations. Part three will introduce and explain the
important and novel data used in this study along with the methodology. The results of this study will be discussed in
Section 4. This article ends with a discussion, limitations, and conclusion.

2. Sources and types of innovation

2.1 Sources of innovations
Several studies and reports have addressed the issue of where ideas come from and the sources/actors that contribute to
innovation (e.g. Arundel, 2001; Borins, 2001; Arundel and Geuna, 2004; OECD/Eurostat, 2005, 2018; Walker, 2006;
Hu¨sig and Mann, 2010; APSC, 2011; Bloch and Bugge, 2013; Torugsa and Arundel, 2016b; Demircioglu and Audretsch,
2017b). For instance, Arundel (2001) classified sources of innovation as external (e.g. customers, other firms, and their
parent firm) and internal (e.g. from in-house R&D activities) with using the 1992 Community Innovation Survey.
Analyzing surveys from 1993 Policies, Appropriation, and Competitiveness in Europe (PACE) survey, 1994 Carnegie
Mellon Survey, and 1997 Community Innovation Survey II, Arundel and Geuna (2004) have identified public research
institutes and universities, affiliated firms, suppliers, and customers as important sources of innovation. However, these
major studies did not analyze workers and as a source of innovation.
Additionally, although these studies and our study have some similarities because both focus on sources or types of
innovations, many differences exist. First, our dependent variable is type of innovation and overall innovations, while most
other studies focus on the amount of R&D expenditures, patent applications, location, and geographic proximity of
knowledge sources. Second, data for major studies focusing on sources of innovation are mostly from the 1990s and are not
current. Third, other studies’ sample size is usually small [e.g. Arundel and Geuna’s (2004) main dataset, PACE, has only
588 responses]. Fourth, these many studies cover only the largest R&D-performing firms, so they generally exclude smaller
firms and non R&D firms, which is also true for the US Business R&D and the Innovation Survey of 2008. Our study, on
the other hand, is not restricted to high-performing firms. Fifth, most of the studies on innovation focus on Europe, but the
focus of this study is on the United States.
Several studies analyzing public sector organizations have found that employees, regardless of their rank is very
important and provides a valuable source of innovation in the public sector (Borins, 2001, 2002; APSC, 2011; Bloch and
Bugge, 2013; Torugsa and Arundel, 2016b; Bugge and Bloch, 2016; Demircioglu and Audretsch, 2018). However, one of the
greatest limitations of these empirical and case studies is that the findings are limited only to public sector organizations and
exclude private firms. Additionally, due to data limitations, studies using the Australian Public Service Commission (APSC)
1368 M. A. Demircioglu et al.
data have focused only on individual and group levels of analysis and have generally ignored the knowledge sources and
types of innovation at the organizational level. In contrast, the data used in this article make it possible to include both
workers and universities as sources of innovation at the firm level. This is a very important and highly valued contribution to
the literature. As Nakamura has pointed out, workers provide important sources of creativity and knowledge fueling firm
innovation in the United States on the basis of his findings that the professional creative workers accounted for 0.7% of all
employment (0.2 million) in 1900 and 1.9% of all employment (1.1 million) in 1950, and increased to 5.7% in 1999 (7.6

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
million), suggesting that the number of workers involved in creative and innovation activities continues to increase
(Nakamura, 2000).
Additionally, Arundel and Geuna (2004) summarized many studies on innovation and patent applications and found that
in both North America and Western Europe, knowledge and ideas emanating from universities and research institutions
increase the quality and quantity of innovations. When firms are located far away from universities and R&D laboratories,
firm-university interactions are less frequent, and firms may receive fewer ideas, ultimately reducing their innovation
activity (Adams, 2001). In a recent study, Richardson et al. (2016: 132) find that “both scholars and public policy makers
have viewed investment in university research as a key component to generating innovation activity.” Thus, the effects of
universities on types of innovation and overall innovations are vital to analyze. As Audretsch and Stephan (1996) find,
“companies can also learn about the research occurring in university labs through social contacts between employees and
university scientists and by sending employees to participate in workshops and seminars at the university” (Audretsch and
Stephan, 1996: 646). Furthermore, Demircioglu and Audretsch (2017b) find that ideas emanating from universities are
positively associated with the quality of public services, employee job satisfaction, and interagency collaboration. Thus, we
expect that ideas emanating from workers and universities are positively associated with types and overall innovations and
the association is statistically significant and meaningful.

2.2 Types of innovations

As the Oslo Manual argues, “It is not enough to know whether firms are innovative or not; it is necessary to know how
firms innovate and what types of innovations they implement” (OECD/Eurostat, 2005: 13). Innovations can be defined
differently. Slaper et al., use the following definitions: “According to the U.S. Department of Commerce (2008), innovation
is ‘the design, invention, development and/or implementation of new or altered products, services, processes, systems,
organizational structures, or business models for the purpose of creating new value for customers and financial returns for
the firm.’ Other authors offer similar definitions, including the Federal Reserve Bank of St. Louis (2007), which defines
innovation as ’taking something established and introducing a new idea, method or device that creates a new dimension of
performancea˜ and adding value’" (Slaper et al., 2011, 37). According to Arundel (2001: 614), a firm is “innovative if it
introduced a product or process innovation... includes firms that only make minor improvements or introduce new products
and processes developed by other firms. Many of these ‘minor’ innovators might not develop patentable innovations.”
Therefore, innovations are different than inventions and the focus of this study is innovation.
Innovation is typically multidimensional (Torugsa and Arundel, 2016a; Demircioglu and Audretsch, 2018), so that
several types of innovation exist. There are a number of distinct types of innovation activity, such as product, process, and
organizational innovation (OECD/Eurostat, 2005, 2018; Walker, 2006; Radicic et al., 2016; Gault, 2018). The Oslo manual
was developed by the OECD and Eurostat in an effort to measure and analyze the impact of innovation activity
(OECD/Eurostat, 2005, 2018). According to the OECD, product innovation has two dimensions—goods and services
innovation, both of which introduce a “good or service that is new or significantly improved with respect to its
characteristics or intended uses” (OECD/Eurostat, 2005: 48). Process innovation is “the implementation of a new or
significantly improved production or delivery method [e.g. techniques or software]” (OECD/Eurostat, 2005: 49), and
marketing innovation is “the implementation of a new marketing method involving significant changes in product design or
packaging, product placement, product promotion or pricing” (OECD/Eurostat, 2005: 49).
All of these three types of innovations are crucial for organizations. For example, product innovation leads to an
increase in firm profitability, as buyers tend to buy a new or updated product with more cost and process innovations
usually leads improvement of methods and reduce costs (Fritsch and Meschede, 2001). Process innovations can create
competitive advantage, efficiency improvements, and product growth to firms, justify government support for innovations,
and help for economic growth (Terjesen and Patel, 2017). Marketing innovations can increase organizational survival
particularly during economic crises and will increase their competitive advantage (e.g. reducing costs of their products,
increasing differentiation, and increasing their focus; Naidoo, 2010; see also Gault, 2018; OECD/ Eurostat, 2018).
Sources of innovation and innovation type 1369
Examples of a product innovation include “robotic vacuum-cleaners in nursing homes for the elderly; environmentally
friendly ambulances; and the use of the ‘PARO seal’ in residential homes (the PARO seal is a robot shaped like a seal, which
is able to move its eyes and make sounds)” (Bugge and Bloch, 2016: 286). Examples of process innovations include “a new
IT solution for joint login to public services, which enabled switching between services without new logins for each service;
and self-service in the personnel and payroll system” (Bugge and Bloch, 2016: 286) and “to cut down decision layers from
five to just two, reducing approval time by 80 per cent” (Van der Wal, 2017: 170). An example of a marketing innovation is

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
the creation of online stores, even while retail companies continue doing their regular business at their psychical location,
such as Amazon.com’s “one-click” ordering process at online (Chen, 2006).

3. Data and methods

3.1 Data and sample
Data come from the National Survey of Business Competitiveness (NSBC), which was conducted by the Social &
Economic Sciences Research Center (SESRC) at Washington State University for the United States Department of
Agriculture’s Economic Research Service. Before distributing the questionnaire, pilot study data collection started in
November 2013 and December 2013. After the pilot study, over 53,000 respondents were invited to complete the surveys
via (i) online, (ii) phone interview, or (iii) mail from April through November 2014. All respondents received an invitation
letter, questionnaire, reminder (by a postcard), questionnaire again, and phone and email reminders, respectively. About
10,952 respondents (22.4%) at least partially completed the survey. There are no statistical differences between respondents
and nonrespondents (USDA, no date, Part C). In addition, no significant differences exist among different types of survey
completion methods. Finally, the final sample size was reduced to 4845 due to missing variables and incomplete parts of
the responses. There are no systematic differences between missing and without missing variables.
The population for this study is business establishments in the United States; businesses were selected randomly from
strata based on industry codes (NAICS), location (metropolitan vs. nonmetropolitan area), and establishment size. The
sample of this survey has been chosen from the business establishment list maintained by the Bureau of Labor Statistics.
Washington DC and 46 out of 50 states have participated to the survey (Wojan and Parker, 2017). Additionally, NSBC is
the first national and representative survey on innovation in the United States. 1
Some of the questions of NSBC, particularly types of innovation, come from the European Union’s Community
Innovation Survey (EU-CIS), with the hope that future researchers compare innovation activities between European
countries and the United States (USDA, no date, Part A). Although the first national level surveys on innovation in the
United States was Business R&D and the Innovation Survey of 2008, these data are limited to only larger firms,

1 “The confidentiality of the Rural Establishment Innovation Survey data is protected under the statutes of US Code Title
18, Section 1905, US Code Title 7, Section 2276, and Title V of the E-Government Act, Confidential Information
Protection and Statistical Efficiency Act of 2002 (CIPSEA; Public Law 107-347). A full statement is provided in mail
questionnaires and web surveys confirming that answers to the survey will be kept confidential, and that under no
circumstances will identifying information about individuals or their organizations be released to any unauthorized
individuals, agencies, or institutions. It will assure respondents that only aggregated statistics will be reported, and
that providing answers to any or all questions is strictly voluntary. Detailed disclosures regarding confidentiality will
be provided in an advance letter to respondents (see Attachment D), and enumerators will check to ensure that
respondents have received and read the letter and disclosures prior to conducting the survey. Economic Research
Service will use established procedures for survey storage and disposal to ensure that individual identifiers are
protected from disclosure. ERS will also use statistical disclosure limitation methods to ensure that individual
identifying information does not appear in any public data product” (United States Department of Agriculture, 2014,
Part A: 14).
1370 M. A. Demircioglu et al.

Table 1. Descriptive statistics

Variable Mean SD Min. Max.

Product innovation 0.58 0.41 0 1

Process innovation 0.45 0.39 0 1

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
Marketing innovation 0.46 0.50 0 1
Source 1: Suppliers 2.26 0.64 1 3
Source 2: Customers 2.40 0.64 1 3
Source 3: Others in industry 2.38 0.63 1 3
Source 4: Workers 2.35 0.63 1 3
Source 5: Universities 1.55 0.65 1 3
Establishment yearsa 9.22 3.32 1 14
Number of employeesb 8.33 4.11 1 15
Rurality index (Beale13) 4.87 2.28 1 9
Industry Classification (Naics2)
21: Mining 0.26 0.16 0 1
31: Food manufacturing 0.08 0.27 0 1
32: Wood, paper manufacturing 0.13 0.33 0 1
33: Metal, machinery, electronic, hardware 0.21 0.40 0 1
42: Wholesale trade 0.20 0.40 0 1
48: Transportation 0.06 0.24 0 1
51: Information 0.07 0.25 0 1
52: Finance and insurance 0.02 0.15 0 1
54: Professional 0.17 0.37 0 1
55: Management of companies 0.02 0.15 0 1
71: Arts, entertainment 0.02 0.14 0 1
n ¼ 4845
a
Years categories: 1800–1871¼1, 18721896¼2, 1897–1921¼3, 1922–1946¼4, 1947–1962¼5, 1963–1972¼6, 1973–1977¼7, 1978–1982¼8, 1983– 1987¼9, 1988–1992¼10,
1993–1997¼11, 1998–2002¼12, 2003–2007¼13, 2008–2012¼14.
b
Employment categories: Between 1 and 4 employees¼1, 5¼2, 6¼3, 7¼4, 8¼5, 9¼6, 10¼7, 11–12¼8, 1–15¼9 16–19¼10 20–29¼11, 30–49¼12, 50– 99¼13, 100–249¼14,
250–499¼15, 500–14,000¼15.

so they have a very limited focus. NSBC, on the other hand, is very unique in its broad representation. This project is
consistent with the White House’s emphasis on innovation. For instance, President Obama’s 2011 State of the Union
focused on innovation in America, including both urban and rural areas, as they key to national prosperity in the United
States (USDA, no date, Part A).

3.2 Dependent variables

Four different dependent variables are used in this article. The first three dependent variables reflect the type of innovation,
namely whether the business participated in (i) product (including goods and service innovations), (ii) process, and (iii)
marketing innovations. All variables used in this study come from the respondents who represent their organizations. Thus,
the levels of analysis are the firm. The first dependent variable asks whether an organization produced any new or
significantly improved products. Two survey items capture the product innovation—whether the business has involved
goods and service innovation (OECD/Eurostat, 2005, 2018). Appendix 1 (Table A1) shows all the operationalization used
in this study.
The second dependent variable asked whether the firm introduced new or significantly improved methods of
manufacturing or producing goods or services, introduced new or significantly improved logistics, delivery, or distribution
methods for the firm’s inputs, goods, or services, and introduced new or significantly improved support activities for the
firm’s processes. These three items capture process innovations. The third dependent variable asks whether new or
significant improvements in the firm’s marketing methods were implemented to measure marketing innovations. One survey
Sources of innovation and innovation type 1371

item reflects marketing innovations. Finally, the last dependent variable is firm innovation activity, which includes all six
indicators (including product, process, and marketing innovation).
3.3 Independent variables
Five main sources of knowledge, or innovation sources are used as independent variables. Important sources or new
information identified by the National Survey of Business Competitiveness (2014) are: (i) suppliers, (ii) customers, (iii) other
people in industry, (iv) workers, and (v) universities. Respondents were asked to evaluate how those sources of new

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
information are valuable for the firm. The response categories are “not at all valuable,” “somewhat valuable,” and “very
valuable.” These identified sources are mostly consistent with Australian Public Service Commission’s 2011 data which
important sources are identified (APSC, 2011), Borins’s analysis of sources of innovation (2011), the Oslo Manual (2018),
and Terjesen and Patels’ research on process innovations (2017).

3.4 Control variables

Because type, size, location, and establishment year of firm have been consistently identified in the extant literature as
influencing innovation activity (e.g. Damanpour, 1991), they need to be included in the regression analysis as control
variables. For example, analyzing the innovation activities of Europe’s major industrial organizations of CIS, Arundel and
Kabla found that, on average, firm product innovation rates in the pharmaceutical industry is approximately 80%, while in
textiles it is only 8%. Similarly, Ballot et al. (2015) found that the particular types of industry (e.g. food manufacturing,
heavy machine manufacturing, transportation, information, and finance and insurance companies) can affect innovation
activity.
In addition, the size of a firm affects innovative performance in that large organizations tend to perform better thanks to
their larger budgets and personnel (Arundel, 2001; Lœgreid et al., 2011; Ballot et al., 2015). Moreover, the location of the
firms (e.g. their location in either an urban or rural area) and the number of employees can similarly effect innovation
(Arundel and Kabla, 1998; Carlino, 2001). In this study, we have used Beale codes, which is a Rural-Urban Continuum
Code (between 1¼more urban and 9¼more rural) developed by the United States Department of Agriculture (USDA) to
reflect the geographic location. This nine-scale variable is better than a binary metro versus non-metro variable. Finally,
firm age affects innovations because on the one hand, older firms may have more established innovation routines, so they
are more innovative (Terjesen and Patel, 2017). On the other hand, new organizations “will have less fixed identities, norms
and values, enabling them to think ‘outside the box,’” and thus they tend to be more innovative (Lœgreid, et al., 2011:
1328). Thus, we have also controlled these factors. Table 1 shows descriptive statistics of all the variables used in this study
Appendix 2 also reports correlation coefficients.

3.5 Estimation
As explained in the above section, the set of dependent variables reflects the type of innovation, namely product, process,
and marketing innovation. Because these three dependent variables are correlated with each other (error terms are
correlated), using ordinary least square (OLS) or ordinal logit models cause inefficient results, since the latter models
assume that the errors are not correlated with each other (Zellner, 1962; Fiebig, 2001; Martin and Smith, 2005). For
instance, the correlation between product and process innovation is 0.56, the correlation between product and marketing
innovation is 0.43, and the correlation between process and marketing innovation is 0.8, indicating that the errors are
somewhat correlated. To compensate for this problem, seemingly unrelated regressions (SUR) are used ( Zellner, 1962;
Fiebig, 2001; Martin and Smith, 2005). Stata software program can provide results of the SUR precisely and efficiently.
OLS is used for the fourth and last dependent variable—overall firm innovation activity as this variable is a summated scale
of six indicators. The next section will report and summarize the results.

4. Results
Table 1 reports the descriptive statistics. On average, 58% of companies reported to have implemented product innovation,
46% of companies reported to have implemented marketing innovation, and 45% of companies reported to have
implemented process innovation. The highest mean value of the knowledge sources is customers, suggesting that customers
involve many of the innovation activities. There is considerable variance in establishment year in the survey, so both very
new and old firms exist, which also confirms the representativeness of the survey. Results of
1372 M. A. Demircioglu et al.

Table 2. Regression results

Sources of innovation and innovation type 1373
Variable Product innovation Process innovation Marketing innovation All innovations

Source 1: Suppliers 0.023* 0.013 0.006 0.014

Source 2: Customers 0.061*** 0.053*** 0.054*** 0.056***
Source 3: Others in industry 0.01 0.020* 0.033** 0.015
Source 4: Workers 0.043*** 0.052*** 0.053*** 0.049***

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
Source 5: Universities 0.074*** 0.076*** 0.096*** 0.082***
Establishment years 0.010*** 0.009*** 0.010*** 0.010***
Number of employees 0.010*** 0.015*** 0.018*** 0.014***
Rurality index (Beale13) 0.009*** 0.009*** 0.011*** 0.010***
Industry classification (Naics2)Base: 21: mining
31: Food manufacturing 0.194*** 0.098** 0.037 0.110**
32: Wood, paper manufacturing 0.133*** 0.067 0.025 0.075*
33: Metal, machinery, electronic, hardware 0.239*** 0.117*** 0.089* 0.149***
42: Wholesale trade 0.140*** 0.080* 0.121** 0.113***
48: Transportation 0.052 0.015 0.001 0.022
51: Information 0.306*** 0.175*** 0.183*** 0.221***
52: Finance and insurance 0.176*** 0.087 0.216*** 0.160***
54: Professional 0.125*** 0.048 0.117* 0.096**
55: Management of companies 0.071 0.016 0.065 0.04
71: Arts, entertainment 0.283*** 0.098* 0.166* 0.183***
Constant 0.102 0.237*** 0.319*** 0.219***
R2 0.11 0.1 0.08 0.12
*P<0.05; **P<0.01; ***P<0.001; n¼4845.

the industry classification (Naics2) show that most of the firms in the sample are in the following two-digit sectors: 21:
mining, 33: metal, machinery, electronic, and hardware, and 42: wholesale trade, followed by 54: professional jobs and 32:
wood and paper manufacturing.
Table 2 reports the results of the unstandardized SUR models. Accordingly, ideas emanating from customers, workers,
and universities are positively associated with all types of innovations (P<0.001). For instance, ideas emanating from
customers are positively associated with namely product (B¼0.06, P<0.001), process (B¼0.05, P<0.001), and marketing
innovations (B¼0.05, P<0.001), holding other variables constant. Ideas emanating from suppliers are moderately and
positively associated with only product innovation (B¼0.02, P<0.05), but suppliers do not have any statistical relationship to
other types of innovations. Ideas emanating from other business people in the firm’s industry are positively associated with
process and marketing innovations (P<0.05), holding other variables constant, although the magnitude of the effect is small.
Ideas emanating from workers are positively associated with all types of innovations (product, process, and marketing),
holding other variables constant (B¼0.04, 0.05, and 0.05, respectively; P<0.001). Finally, the universities as sources of
innovation are also positively and statistically associated with both types of innovations ( P<0.001). In fact, the university as
a source of innovation has higher coefficients (B¼0.07, 0.08, and 0.1, respectively), suggesting that universities are
important sources of innovation.
Regarding the control variables, on average, the younger firms, firms with more employees, and firms located in urban
areas exhibit a greater degree of innovation activity for all types of innovations—product, process, and marketing
innovations. All of these results are statistically significant at the 0.01% level. In addition, compared with other industry
types, companies in information (Naics¼51), metal, machinery, electronic, hardware (Naics¼33), and arts and
entertainment (Naics¼71) are more innovative than are other companies. Moreover, on average, companies tend to generate
more product innovations compared with process innovations. Finally, the results show that transportation firms and
management of companies do not exhibit a significant amount of innovation activity suggesting that the particular type of
organization matters for innovation.
1374 M. A. Demircioglu et al.

5. Discussion
Innovation, which is one of the most important topics in the studies of business, economics, and management can contribute
to organizational performance, in that is conducive to organizational survival, success such as increasing competitiveness,
productivity, and performance (e.g. Burns and Stalker, 1961; OECD/Eurostat, 2005, 2018; Pianta, 2006; Damanpour et al.,
2009; Frenz and Ietto-Gillies, 2009; Thompson, 2010; Stojcic and Hashi, 2014; Demircioglu and Audretsch, 2017a; Wojan

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
et al., 2018). The extent, scale, and scope of the studies on innovation and innovation activities reflect the prominence of
innovation (e.g. Mohr, 1969; Kimberly and Evanisko, 1981; Damanpour, 1991; Greenhalgh et al., 2004; Smith, 2006;
Ballot et al., 2015; Torugsa and Arundel, 2016a, 2016b; – >Demircioglu and Audretsch, 2018; Kattel and Mazzucato,
2018; Mazzucato, 2018). Analyzing how five important sources of innovation (i.e. suppliers, customers, others in industry,
workers, and university) are associated with types of innovation (i.e. product, process, and marketing), we have found that
universities, workers, and customers are important sources of innovation positively associated with all of innovation types.
This article has made several unique and important contributions that are discussed in below.
This main focus of this article is sources of innovation. Although identifying sources of innovation goes back to the late
19th century (Carlino, 2001), testing for the effects of different knowledge sources and relating them to different outcomes
is a relatively recent topic. Using ideas from different knowledge sources are crucial for generating different types of
innovations and outcomes (Audretsch and Stephan, 1996; Arundel and Geuna, 2004; Frenz and Ietto-Gillies, 2009).
Terjesen and Patel (2017: 1422) suggest that organizational managers must recognize and take advantage of ideas
emanating from a broad spectrum of knowledge sources (e.g. universities) because “this knowledge is subsequently stored
in the organization’s memory (i.e., equipment, systems, procedures, routines, and managers) and becomes ‘organizational
knowledge’ that can be harnessed to enhance a firm’s innovation activities... [which are] imperative to innovation.” The
empirical results of this article show that particular knowledge, such as customers, workers, and universities can enable
firms to generate product, process, and marking innovations. Knowledge and ideas emanating from customers positively
associated all types of innovation, which is consistent with the hypothesis that feedback and ideas generated by customers
enable organizations to become more innovative.
While workers provide internal sources of knowledge within the organization or firm, universities and customers
provide external sources of innovations. The results imply that both internal and external sources are necessary to develop
different types of innovations. Moreover, as the positive and statistically significant coefficients of variables measuring
universities and workers indicate, the results support the view that universities and workers are key sources of ideas and
innovations. This is particularly important from the perspective of organizational managers, who should encourage their
workers to express their views and facilitate their feedback to enhance innovation activity. Future studies may focus on
analyzing how workers in the organizational context of universities specifically influence different types of innovations.
In addition, rather than focusing only on a sole type of innovation or aggregated innovation activity, the present study
provides and compares across three different types of innovations—product, process, and marketing innovations.
Distinguishing among product, process, and marketing innovations is essential because each type of innovation has distinct
and different characteristics (e.g. process innovations are less tangible), may be influenced by different knowledge sources
(e.g. different external sources are associated with different types of innovations differently), may require different
resources (e.g. process innovations may be more costly in terms of monetary sources), and may lead to different
organizational outcomes (e.g. product innovations may lead to patents compared with other types of innovations; Fritsch
and Meschede, 2001; OECD/Eurostat, 2005, 2018; Chen, 2006; Naidoo, 2010; Stojcic and Hashi, 2014; Radicic et al.,
2016; Terjesen and Patel, 2017; Van der Wal, 2017; Gault, 2018). The results of this study suggest that different types of
knowledge sources are associated with distinct types of innovations differently. Therefore, future studies would be wise to
extend and build on this study by comparing different types of innovations.
Moreover, this study uses a recent (2014), comprehensive (both rural and urban area with almost all states in the United
States), and large-scale survey (n¼10,952 firms) on firm innovation in the United States. Despite many
Sources of innovation and innovation type 1375

datasets in Europe [e.g. Community Innovation Surveys (CIS)], fewer datasets on firms’ innovation activities are available
for the United States. This is unfortunate because many large and innovative companies such as Apple, Uber, Amazon,
Netflix, and Microsoft are located in the United States. Furthermore, the United States ranks among the large advanced
countries in the world, with a population of over 320 million. Therefore, a key contribution of this study is analyzing survey
responses from the National Survey of Business Competitiveness (NSBC) which is sponsored by the United States

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
Department of Agriculture. The findings provide interesting results, such as the more robust innovation activity exhibited
by firms in the urban areas, but also both the extent and heterogeneity of innovation activity exhibited by firms in rural
areas. This article also shows that on average, certain types of industries (e.g. IT, metal, machinery, electronic, and arts) are
more innovative than other industries. Additionally, both firm age and size are positively associated with innovation types.
All of these findings made possible by analyzing these new and unique data provide insights that will be instructive and
enlightening for decision makers in both business and policy.
Several important limitations inherent in this study need to be emphasized. Like most of survey research, this article is
based on a cross-sectional data (United States Department of Agriculture (USDA), Economic Research Service, 2014).
Therefore, lack of panel data, or quasi-experimental design does not permit scholars to make causal claims about findings.
In other words, the study is correlational, not causal. In addition, due to data limitations, we could not measure which
particular innovations are more important and effective than others. For example, data does not provide information about
the quality of innovations such as which innovations lead huge savings, sales, or patents. Similarly, the outcomes variables
are only probably to innovate in a given, so the outcome is same if a given firm introduces only one innovation versus five
innovations in the previous three years. This limitation (both quality—the significance of innovation—and the quantity—
the number of innovations–) was also acknowledged from the USDA as it was stated as “a noted weaknesses of this set of
questions is their inability to differentiate highly innovative from nominally innovative establishments” (USDA, Part A: 4).
Thus, future research may focus on the quality and quantity of innovation with using more questionnaires. Future research
may also use qualitative methods (e.g. cases, in-depth interviews, and ethnographic research) to examine how and why
certain sources of innovative ideas lead major innovations because qualitative research are important to explain how and
why a particular relationship occurs, not occurs, or occur in different intensity (Corbin and Strauss, 2008; Maxwell, 2012;
Bryman, 2016).
Despite the limitations of this study, it offers insights and practical implications about how sources of innovations are
associated with types of innovations, the importance of workers, customers, and universities as innovation sources, the
importance of product, process, and marketing innovations. We recommend future studies to focus on the link of sources
and types of innovation in other context with qualitative research. Future research may also focus on how and why
universities, customers, and employees influence innovation activities and benefits of innovations (e.g. whether innovation
from particular sources reduces cost and increases quality of produce and services) in public, nonprofit, and private settings.
Future research may also explore the effects of other sources of knowledge such as the Federal government’s role for
innovation activities of companies and public organizations.

6. Conclusion
Innovation has emerged not just as an important force driving the performance of firms and other organizations, but also
entire regions and countries. Despite its prominence, there are not many quantitative studies analyzing how different
knowledge sources are associated with different types of innovations as well as overall innovation activity. Using United
States National Survey of Business Competitiveness (NSBC) conducted in 2014, this study has contributed to the
innovation literature by analyzing how the following knowledge sources are associated with product, process, and
marketing innovation as well as overall innovation activity by the following important innovation sources: (i) suppliers, (ii)
customers, (iii) other people in industry, (iv) workers, and (v) university. The most important findings of this article are that
universities, customers, and workers are crucial knowledge sources, positively associated with product, process, marketing,
and overall innovation activity. Suppliers and others in industry have either little or no statistical effect for innovation
activities.
Just as knowledge sources are heterogeneous, so too are the manifestations of innovation activity highly heterogeneous.
Yet, the literature on innovation activity has generally ignored the heterogeneity inherent in both knowledge sources and
manifestations of innovation. If there is an innovation, there should be a knowledge source(s), other than the skies to which
Solow eluded driving innovation activity. Our study is original and important to the literature as it has found that
universities, workers, and customers are crucial sources for innovations. With using qualitative or quantitative research,
future studies may analyze other sources (e.g. governments’ role for innovation), other types of innovations (e.g.
1376 M. A. Demircioglu et al.

organizational), or the benefits of innovation (e.g. whether a particular innovation increases efficiency) in other context both
in terms of different region (e.g. in Asia) and in different sector (e.g. public and nonprofit).

Acknowledgment

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
We appreciate the helpful comments and suggestions of Barry M. Rubin (Indiana University-Bloomington), Chantalle LaFontant (Indiana
University-Bloomingon), Stephan Goetz (Pennsylvania State University), and Tim Mulcahy (National Opinion Research Center) along with
the Editors and anonymous reviewers in this journal. In particular, we have received excellent feedback and help from Timothy R. Wojan,
who is a Senior Economist at the Economic Research Service/USDA ever since we started the project. The authors are responsible for any
remaining errors.

Funding
This research was supported in part by the intramural research program of the U.S. Department of Agriculture, Economic Research Service.
Extramural funding for this research provided by the Economic Research Service was through USDA/REE Cooperative Agreement 58-
6000-5-0095, administered by the Northeast Regional Center for Rural Development (NERCRD). The research for this paper was also
financially supported by National University of Singapore (NUS) (Grant No. R-603-000-270-133) and Indiana Business Research Center
(IBRC) at Indiana University-Bloomington. The views expressed are those of the authors and should not be attributed to the Economic
Research Service, the USDA, NERCRD, NUS, or IBRC.

References
Aboal, D. and E. Tacsir (2018), ‘Innovation and productivity in services and manufacturing: the role of ICT,’ Industrial and Corporate
Change, 27(2), 221–241.
Acs, Z. J., D. B. Audretsch, E. E. Lehmann and G. Licht (2017), ‘National systems of innovation,’ The Journal of Technology Transfer,
42(5), 997–1008.
Aghion, P. and P. Howitt (1998), Endogenous Growth Theory. MIT Press: Cambridge, MA.
Arundel, A. (2001), ‘The relative effectiveness of patents and secrecy for appropriation,’ Research Policy, 30(4), 611–624.
Arundel, A. and A. Geuna (2004), ‘Proximity and the use of public science by innovative European firms,’ Economics of Innovation and
New Technology, 13(6), 559–580.
Arundel, A. and I. Kabla (1998), ‘What percentage of innovations are patented? Empirical estimates for European firms,’ Research Policy,
27(2), 127–141.
Audretsch, D. B. and P. E. Stephan (1996), ‘Company-scientist locational links: the case of biotechnology,’ The American Economic
Review, 86(3), 641–652.
Australian Public Service Commission (APSC) (ed) (2011), Employee Survey Results: State of the Service Series 2010–11. Australian
Public Service Commission: Canberra, Australia.
Ballot, G., F. Fakhfakh, F. Galia and A. Salter (2015), ‘The fateful triangle: complementarities in performance between product, process and
organizational innovation in France and the UK,’ Research Policy, 44(1), 217–232.
Bankins, S., B. Denness, A. Kriz and C. Molloy (2017), ‘Innovation agents in the public sector: applying champion and promotor theory to
explore innovation in the Australian public service,’ Australian Journal of Public Administration, 76(1), 122–137.
Battisti, G. and P. Stoneman (2010), ‘How innovative are UK firms? Evidence from the fourth UK community innovation survey on
synergies between technological and organizational innovations,’ British Journal of Management, 21(1), 187–206.
Bloch, C. and M. M. Bugge (2013), ‘Public sector innovation—from theory to measurement,’ Structural Change and Economic Dynamics,
27, 133–145.
Birkland, T. A. (2011), An Introduction to the Policy Process: Theories, Concepts and Models of Public Policy Making. Routledge: New
York, NY.
Borins, S. (2001), The Challenge of Innovating in Government. PricewaterhouseCoopers Endowment for the Business of Government:
Arlington, VA.
Borins, S. (2002), ‘Leadership and innovation in the public sector,’ Leadership & Organization Development Journal, 23(8), 467–476.
Breznitz, D., D. Ornston and S. Samford (2018), ‘Mission critical: the ends, means, and design of innovation agencies,’ Industrial and
Corporate Change, 27(5), 883–896.
Brown, K. and S. P. Osborne (2012), Managing Change and Innovation in Public Service Organizations. Routledge: New York, NY.
Bryman, A. (2016), Social Research Methods. 5th edn. Oxford University Press: Oxford, UK.
Bugge, M. M. and C. W. Bloch (2016), ‘Between bricolage and breakthroughs—framing the many faces of public sector innovation,’ Public
Money & Management, 36(4), 281–288.
Sources of innovation and innovation type 1377

Burns, T. E. and G. M. Stalker (1961), The Management of Innovation. Tavistock: London.

Carlino, G. A. (2001), ‘Knowledge spillovers: cities’ role in the new economy,’ Business Review, Q4, 17–24.
Cantwell, J. (2006), ‘Innovation and competitiveness,’ in J. Fagerberg and D. C. Mowery (eds), The Oxford Handbook of Innovation. Oxford
University Press: New York, pp. 543–567.
Chen, Y. (2006), ‘Marketing innovation,’ Journal of Economics & Management Strategy, 15(1), 101–123.

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
Cooper, R. S. and S. A. Merrill (eds) (1997), Industrial Research and Innovation Indicators: Report of a Workshop. National Academies
Press: Chicago, IL.
Corbin, J. and A. Strauss (2008), Basics of Qualitative Research. 3rd edn. Sage: Thousand Oaks, CA.
Damanpour, F. (1991), ‘Organizational innovation: a meta-analysis of effects of determinants and moderators,’ Academy of Management
Journal, 34(3), 555–590.
Damanpour, F., R. M. Walker and C. N. Avellaneda (2009), ‘Combinative effects of innovation types and organizational performance: a
longitudinal study of service organizations,’ Journal of Management Studies, 46(4), 650–675.
Demircioglu, M. A. and D. B. Audretsch (2017a), ‘Conditions for innovation in public sector organizations,’ Research Policy, 46(9), 1681–
1691.
Demircioglu M. A. and D. B. Audretsch (2017b), ‘Public sector innovation: the effect of universities,’ The Journal of Technology Transfer,
1–19.
Demircioglu, M. A. and D. B. Audretsch (2018), ‘Conditions for complex innovations: evidence from public organizations,’ The Journal of
Technology Transfer, 1–24.
Fagerberg, J. and M. M. Godinho (2006), ‘Innovation and catching-up,’ in J. Fagerberg and D. C. Mowery (eds), The Oxford Handbook of
Innovation. Oxford University Press: New York, NY, pp. 514–542.
Fiebig, D. G. (2001), ‘Seemingly unrelated regression,” in B. H. Baltagi (ed), A Companion to Theoretical Econometrics. Blackwell
Publishing Ltd: Malden, MA, pp. 101–121.
Frenz, M. and G. Ietto-Gillies (2009), ‘The impact on innovation performance of different sources of knowledge: evidence from the UK
Community Innovation Survey,’ Research Policy, 38(7), 1125–1135.
Fritsch, M. and M. Meschede (2001), ‘Product innovation, process innovation, and size,’ Review of Industrial Organization, 19(3), 335–350.
Gault, F. (2018), ‘Defining and measuring innovation in all sectors of the economy,’ Research Policy, 47(3), 617–622.
Greenhalgh, T., G. Robert, F. Macfarlane, P. Bate and O. Kyriakidou (2004), ‘Diffusion of innovations in service organizations: systematic
review and recommendations,’ Milbank Quarterly, 82(4), 581–629.
Griffith, R., E. Huergo, J. Mairesse and B. Peters (2006), ‘Innovation and productivity across four European countries,’ Oxford Review of
Economic Policy, 22(4), 483–498.
Hu¨sig, S. and H.-G. Mann (2010), ‘The role of promoters in effecting innovation in higher education institutions,’ Innovation, 12(2), 180–
191.
Kattel, R. and M. Mazzucato (2018), ‘Mission-oriented innovation policy and dynamic capabilities in the public sector,’ Industrial and
Corporate Change, 27(5), 787–801.
Kimberly, J. R. and M. J. Evanisko (1981), ‘Organizational innovation: the influence of individual, organizational, and contextual factors on
hospital adoption of technological and administrative innovations,’ Academy of Management Journal, 24(4), 689–713.
Leyden, D. P. and A. N. Link (1992), Government’s Role in Innovation: Springer Science & Business Media: Norwell, MA.
Link, A. N. (2013), Public Support of Innovation in Entrepreneurial Firms. Edward Elgar Publishing: Northampton, MA.
Link, A. N. and D. S. Siegel (2007), Innovation, entrepreneurship, and Technological Change. Oxford University Press: New York.
Lœgreid, P., P., Per, G. Roness and K. Verhoest (2011), ‘Explaining the innovative culture and activities of state agencies,’ Organization
Studies, 32(10), 1321–1347.
Martin, S. and P. C. Smith (2005), ‘Multiple public service performance indicators: toward an integrated statistical approach,’ Journal of
Public Administration Research and Theory, 15(4), 599–613.
Maxwell, J. A. (2012), Qualitative Research Design: An Interactive Approach. Vol. 41. Sage Publications: Thousand Oaks, CA.
Mazzucato, M. (2018), ‘Mission-oriented innovation policies: challenges and opportunities,’ Industrial and Corporate Change, 27(5), 803–
815.
Mohr, L. B. (1969), ‘Determinants of innovation in organizations,’ American Political Science Review, 63(1), 111–126.
Naidoo, V. (2010), ‘Firm survival through a crisis: the influence of market orientation, marketing innovation and business strategy,’
Industrial Marketing Management, 39(8), 1311–1320.
Nakamura, L. (2000), ‘Economics and the new economy: the invisible hand meets creative destruction,’ Business Review, 15–30.
Nasi, G., M. Cucciniello, V. Mele, G. Valotti, R. Bazurli, H. De Vries, V. Bekkers, L. Tummers, M. Gasco´, T. Ysa, C. Ferna´ndez, A.
Albareda, A. Matei, C. Savulescu, C. Antonie, E. B. Balaceanu, N. Juraj, M. Svidronova´, B. M. Merickova, M. Oviska, V. de Froidcourt,
M. Eymeri-Douzans and E. M. Monthuber (2015), Determinants and barriers of adoption, diffusion and upscaling of ICT-driven social
innovation in the public sector: a comparative study across 6 EU countries. In LIPSE Research Report #5.
OECD. (2009), OECD Reviews of Innovation Policy: Mexico 2009. OECD Publishing: Paris.
1378 M. A. Demircioglu et al.

OECD/Eurostat. (2005), Oslo Manual: Guidelines for Collecting and Interpreting Innovation Data. 3rd edn. OECD: Paris.
OECD/Eurostat. (2018), Oslo Manual 2018: Guidelines for Collecting, Reporting, and Using Data on Innovation. OECD: Paris.
Pianta, M. (2006), ‘Innovation and employment,’ in J. Fagerberg and D. C. Mowery (eds), The Oxford Handbook of Innovation. Oxford
University Press: New York, NY, pp. 568–598.
Radicic, D., G. Pugh, H. Hollanders, R. Wintjes and J. Fairburn (2016), ‘The impact of innovation support programs on small and medium

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
enterprises innovation in traditional manufacturing industries: An evaluation for seven European Union regions’. Environment and
Planning C: Government and Policy, 34(8), 1425–1452.
Richardson, A., D. B. Audretsch, T. Aldridge and V. K. Nadella (2016), ‘Radical and incremental innovation and the role of university
scientist,’ in D. B. Audretsch and A. N. Link (eds) Essays in Public Sector Entrepreneurship. Springer: Switzerland, pp. 131–207.
Slaper, T. F., N. R. Hart, T. J. Hall and M. F. Thompson (2011), ‘The index of innovation: a new tool for regional analysis,’ Economic
Development Quarterly, 25(1), 36–53.
Smith, K. H. (2006), ‘Measuring innovation,’ in J. Fagerberg and D. C. Mowery (eds), The Oxford Handbook of Innovation. Oxford
University Press: New York, NY, pp. 148–177.
Stojcic, N. and I. Hashi (2014), ‘Firm productivity and type of innovation: evidence from the community innovation survey 6,’ Croatian
Economic Survey, 16(2), 121–146.
Terjesen, S. and P. C. Patel (2017), ‘In search of process innovations: the role of search depth, search breadth, and the industry
environment,’ Journal of Management, 43(5), 1421–1446.
Thompson, J. D. (2010), Organizations in Action: Social Science Bases of Administrative Theory. 7th edn. Transaction Publishers: New
Brunswick, NJ.
Torugsa, N. and A. Arundel (2016a), ‘Complexity of innovation in the public sector: a workgroup-level analysis of related factors and
outcomes,’ Public Management Review, 18(3), 392–416.
Torugsa, N. A. and A. Arundel (2016b), ‘The nature and incidence of workgroup innovation in the Australian public sector: evidence from
the Australian 2011 state of the service survey,’ Australian Journal of Public Administration, 75(2), 202–221.
United States Department of Agriculture (USDA). (2014), Economic Research Service. Information Collection Request for Rural
Establishment Innovation Survey (REIS): Supporting Statement. Available at: https://www.reginfo.gov/public/do/ PRAViewDocument?
ref_nbr=201212-0536-001 (12 February 2019, date last accessed).
Van der Wal, Z. (2017), The 21st Century Public Manager. Macmillan Education: London, UK.
Verspagen, B. (2006), ‘Innovation and economic growth,’ in J. Fagerberg and D. C. Mowery (eds), The Oxford Handbook of Innovation.
Oxford University Press: New York, NY, pp. 487–513.
Verhoest, K., B. Verschuere and G. Bouckaert (2007), ‘Pressure, legitimacy, and innovative behavior by public organizations,’ Governance,
20(3), 469–497.
Walker, R. M. (2006), ‘Innovation type and diffusion: an analysis of local government,’ Public Administration, 84(2), 311–335.
Wojan, T. and T. Parker (2017), Innovation in the rural nonfarm economy: its effect on job and earnings growth, 2010-2014. Economic
Research Report 238. U.S. Dept. of Agriculture, Economic Research Service (September).
Wojan, T. R., D. Crown and A. Rupasingha (2018), ‘Varieties of Innovation and Business Survival: Does Pursuit or Incremental or Far-
Ranging Innovation Make Manufacturing Establishments More Resilient?,’ Research Policy, 47(9), 1801–1810.
Zellner, A. (1962), ‘An efficient method of estimating seemingly unrelated regressions and tests for aggregation bias,’ Journal of the
American Statistical Association, 57(298), 348–368.
Appendix 1

Table A1. Operationalization of variables

Downloaded from https://a

Dependent variables The response categories are following: ¼ 0No, 1¼ Yes.

Product innovation (alpha¼ 0.53) – In the past 3 years, did this business produce any new or significantly improved
services?
– In the past 3 years, did this business provide any new or significantly improved services?
Process innovation (alpha
¼ 0.68) – In the past 3 years, did this business introduce new or significantly improved methods
of manufacturing or producing goods or services?
– In the past 3 years, did this business introduce new or significantly improved logistics,
delivery, or distribution methods for your inputs, goods, or services?
– In the past 3 years, did this business introduce new or significantly improved support
activities for your processes?
Marketing innovation – In the past 3 years, did this business introduce new or significant improvements in your
 demic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November 2020
Independent variables The response categories are following:¼ 1Not at all valuable, 2¼ Somewhat valuable,
and 3 ¼ Very valuable.
Source 1 Suppliers
Source 2 Customers
Sources
Sourceof3 innovation and innovationOther
type business people in your industry 1379
Source 4 Workers
Source 5 University extension, community colleges, business schools
Control variables
Location-Beale13 1: Counties in metro areas of 1 million population or more
Appendix 2 2: Counties in metro areas of 250,000 to 1 million population
3: Counties in metro areas of fewer than 250,000 population
4: Urban population of 20,000 or more, adjacent to a metro area
5: Urban population of 20,000 or more, not adjacent to a metro area
6: Urban population of 2500 to 19,999, adjacent to a metro area

2020
Downloaded from https://academic.oup.com/icc/article/28/6/1365/5382024 by National University of Singapore user on 02 November
Table A2. Correlations
7: Urban population of 2500 to 19,999, not adjacent to a metro area
18: Completely
2 rural 3or less than
4 2500 urban5 population,
6 adjacent
7 to a metro
8 area 9 10 11
9: Completely rural or less than 2500 urban population, not adjacent to a metro area
1 Establishment years
Product innovation 1800–1871
1 ¼ 1, 1872–1896¼ 2, 1897–1921¼ 3, 1922–1946¼ 4, 1947–1962¼ 5, 1963–
1972¼ 6, 1973–1977¼ 7, 1978–1982¼ 8, 1983–1987¼ 9, 1988–1992¼ 10, 1993–
2 Process innovation 1997¼ 11,1 1998–2002¼ 12, 2003–2007¼ 13, 2008–2012¼ 14
0.56
Employment categories Between 1–4 employees ¼ 1, 5 ¼ 2, 6 ¼ 3, 7 ¼ 4, 8 ¼ 5, 9 ¼ 6, 10 ¼ 7, 11–12 ¼ 8, 13–
3 Marketing innovation 15¼ 9, 16–19¼
0.43 0.80 10, 20–29
1 ¼ 11, 30–49¼ 12, 50–99 ¼ 13, 100–249¼ 14,250–
499¼ 15, Over500¼ 15.
4 Industry classification
Source 1: Suppliers(Naics2) 21:
0.09Mining,31 ¼ Food 0.07
0.09 manufacturing,
1 32¼ Wood and paper manufacturing,
33¼ Metal, machinery, electronic, and hardware manufacturing,¼42 Wholesale
trade, 48¼ Transportation, 51¼ Information, 52¼ Finance and insurance,
5 Source 2: Customers 0.15 0.16 0.12 0.28 1
54¼ Professional, 55¼ Management of companies, 71 ¼ Arts and entertainment.
6 Source 3: Others in industry 0.04 0.09 0.10 0.20 0.20 1

7 Source 4: Workers 0.12 0.15 0.13 0.16 0.28 0.19 1

8 Source 5: Universities 0.15 0.17 0.16 0.14 0.08 0.17 0.18 1

9 Establishment years 0.05 0.04 0.03 0.01 0.06 0.03 0.02 0.07 1

10 Number of employees 0.12 0.17 0.14 0.04 0.07 0.01 0.02 0.08 0.16 1

11 Rurality index (Beale13) 0.07 0.07 0.07 0.02 0.07 0.00 0.02 0.06 0.11 0.05 1
View publication stats