Survey Response Rates in Family Business Research

• DOI: http://doi.org/10.1111/emre.12375
• Author: Martin R.W. Hiebl,David I. Pielsticker
• Published date: 01 March 2020
• Date: 01 March 2020

Survey Response Rates in Family

Business Research

DAVID I. PIELSTICKER

1

and MARTIN R.W. HIEBL

1,2

1

Chair of Management Accounting and Control, University of Siegen, Germany

2

Institute of Management Control and Consulting, Johannes Kepler University Linz, Austria

Although family business survey research is growing in volume and publication in highly regarded management

journals, wedo not yet have evidence on the usual responserates in this researchand on the factors that impact these

response rates. This paper addresses these voids and finds that the average response rates of family business survey

studies published in seven prominent outlets amount to 21%. We also find that the response rates have declined

significantly over time and that the size of the survey population and the establishment of contacts with survey

addressees before sending out questionnaires are significantly associated with response rates. Such precontacts

and reminders seem less fruitful for family respondents than for non-family respondents.

Keywords: family business; family firm; survey; response rate

Introduction

Various reviewsshow that surveys are the most frequently

applied data-gathering method in empirical family

business studies (e.g., Bird et al., 2002; Benavides-

Velasco et al., 2013; Evertet al., 2016). At the same time,

surveys addressing family firms may feature some

specific challenges. For instance,Wilson et al.(2014)note

that family members are typically reluctant to give

detailed information about their firms. In addition, family

business surveys are often intended for family members,

such as owner-managers, as they are usually expected to

be most knowledgeable about both business and family

dynamics (Wilson et al., 2014). However, from prior

research, we know that surveying top executives such as

owner-managers usually leads to lower response rates

compared to surveying actors situated at lower levels of

corporate hierarchies (e.g., Cycyota and Harrison, 2006;

Anseel et al., 2010; Hiebl and Richter, 2018). As a

consequenceof these specifics of family businesssurveys,

Stamm and Lubinski (2011) assert that response rates are

usually low in family business survey research. Low

response rates can make it more difficult for a paper to

be publishedin highly regarded businessand management

journals because they may indicate threats to the validity

and reliability of the findings extracted from the surveys.

For instance, low response rates may indicate a coverage

bias that occurs when important segments of the survey

population are not adequately represented in the analyzed

sample (Speklé and Widener, 2018). Another problem is

non-response bias, which may be linked to low response

rates when there are significant differences between the

few members of the survey population who have

answered the questionnaire and the more numerous

members of the population who have not (Baruch and

Holtom, 2008).These potential problems can helpexplain

why response rates are an important criterion for

evaluating the rigor of survey studies for many journal

editors (Carl ey-Baxter et al., 2009). In addition to these

general problems associated with low response rates,

highly regarded business and management journals may

be accustomed to much higher response rates in studies

that do not focus on family firms. That is, survey studies

published in highlyranked management journalstypically

feature response rates between 40% and 60% (Baruch,

1999; Baruch and Holtom, 2008; Anseel et al., 2010;

Mellahi and Harris, 2016). To the best of our knowledge,

there is currently no data about such average response

rates for family business survey studies.

In addition, the factors that impact the response rates in

family business survey research are not known, but may

be different for surveys of family firms. As indicated

above, such family business surveys are often targeted

towards owner-managers. These owner-managers or

Correspondence: Martin R.W. Hiebl, Chair of Management Accounting

and Control, University of Siegen, Germany. E-mail martin.hiebl@uni-

siegen.de

DOI: 10.1111/emre.12375

©2019 The Authors

European Management Review published by John Wiley & Sons Ltd on behalf of European Academy of Management

(EURAM)

This is an open accessarticle under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License,

which permitsuse and distribution in any medium,provided the originalwork is properlycited, the use is non-commercial

and no modifications or adaptations are made.

European Management Review, Vol. 17, –, (2020)

327 346

family managers have been found to show idiosyncrasies

in their behavior compared to non-owner managers (e.g.,

Armstrong and Hird, 2009; Woods and Joyce, 2016).

For instance, Armstrong and Hird (2009) find that owner

managers differ intheir cognitive style from other people.

Many response-rate-enhancing techniques, such as

reminders or incentives, address phenomena of social

psychology or social exchange (e.g., Roth and BeVier,

1998; Fan and Yan, 2010). That is, such techniques

usually attemptto make use of frequently found cognitive

traits of individuals to increase response rates. Given that

owner-managers and family managers have been found

to show different cognitive styles, we could theorize that

typical response-rate-enhancing techniques may also

work out differently in family business survey research,

which further requires knowledge about how such

techniques fare in family business survey research.

This paper aims to address these voids. More

specifically,based on a review of survey studies published

in journals oftenfeaturing family business studies,we aim

to (1) identify the averageresponse rates and response rate

trends over time in family business survey research and

(2) identify the factors that are associated with high

response rates.

Our analysis of 126 family business survey studies

contributes to the literature on response rates in

management research (e.g., Baruch, 1999; Baruch and

Holtom, 2008; Anseel et al., 2010; Mellahi and Harris,

2016) in severalways. We add to this literature an analysis

of response rate levels and factors associated with

response rates for the specific and highly economically

relevant case of family firms. Our findings show that the

average responserate in family business research is lower

than the figures reported for response rates in general

management research. Our study also adds factors that

are significantly associated with response rates in family

business research. Among these, our findings reveal a

factor that has so far not been considered in prior meta-

analytic studies on response rates: whether the

respondents belong to the controlling family or not. This

family status of respondents also shows significant

interactions with the response-enhancing techniques of

reminders and incentives. Both types of response-

enhancing techniques are associated with higher response

rates in studies that did not address only members of

controllingfamilies. Thus, our results imply that the effect

of reminders and precontacts on response rates seems to

be contingent on the type of survey addressee. That is,

our study complements prior meta-analytic work on

response rates in management research (e.g., Baruch,

1999; Baruch and Holtom, 2008; Anseel et al., 2010;

Mellahi and Harris, 2016) by adding that respondents

who belong to t he firm’s controlling family seem to be a

specifically challenging and idiosyncratic population in

regard to reaching high response rates. Consequently,

editors and reviewers from general-management journals

may need to adapt their expectations on response rates

when evaluating surveys on family firms –especially

those targeting family respondents. In turn, family

business scholars may benefit from considering our

findings when designing their next survey studies.

Therefore, in general, we hope that these results can give

family business scholars and reviewers examining family

business surveyssome guidance in termsof what response

rates to typically expect from family business survey

studies. In addition, the factors correlated with high

response rates may support family business researchers

in achieving higher response rates when conducting their

subsequent survey studies.

The remainder of this paper is structured as follows. In

the next section, we review the existing literature on

response rates, extract factors known to influence

response rates and develop hypotheses on their

application in family business research. Afterwards, we

detail the methods applied to address our above research

objectives. Afterwards, we present descriptive data on

the response ratesin family business survey research over

time and a regression analysis on the factors associated

with such response rates. In the final section, we discuss

our results and present implications for future survey

research on family firms.

Literature review and hypotheses

Response rates are seen as an important quality criterion

of survey studies because higher response rates can

strengthen the confidence in the representativeness and

validity of data produced by survey studies (Mellahi and

Harris, 2016). Conversely, low response rates can cause

smaller data samples, decrease statistical power, limit the

applicability of sophisticated statistical procedures

(Rogelberg and Stanton, 2007) and raise doubts about

the validity of the sample (The American Association

for Public OpinionResearch, 2016). That is, low response

rates may represent or at least signal potential biases in

survey research, such as non-response bias (Rogelberg

and Luong, 1998).

These reasons may explain why there is continued

interest in response rates and the factors affecting them.

As part of this line of research, prior meta-analytical

studies have identified various factors that may influence

response rates in survey research (e.g., Roth and BeVier,

1998; Baruch, 1999; Edwards et al., 2002; Cycyota and

Harrison, 2006; Baruch and Holtom, 2008; Anseel et al.,

2010; Chidlow et al., 2015; Mellahi and Harris, 2016;

Hiebl and Richter, 2018). Guidance on achieving high

response rates is also offered in several books (e.g.,

Dillman et al., 2014) and in publications by professional

associations (e.g., The American Association for Public

Opinion Research, 2016). All this work has, however,

D.I. Pielstickerand M.R.W. Hiebl

©2019 The Authors

European Management Review published by John Wiley & Sons Ltd on behalf of European Academy of Management

(EURAM)

328