Research Methods in IB – Trends and Future Agenda

Reports on an AIB 2021 conference panel on research methods in International Business

Jelena Cerar - Sun 3 Oct 2021 07:43 (updated Sat 14 Sep 2024 06:46)

[Guest post by CYGNA member Jelena Cerar. In this post Jelena reports on an AIB 2021 conference panel on research methods in IB that she co-organized]

Since international business has established itself as a research field, we experienced substantial achievements in the form of theoretical advances, construct measurement, and increased legitimacy in the eyes of both scholars and business professionals. Additionally, new data, methods and software that support quantitative and qualitative analysis have become available. Taking all this into account and marking the 50-year anniversary of the Journal of International Business Studies, the time was right to discuss how the methodologies used in IB have developed and what our future agenda as a field ought to be.

This panel – co-organized by Philip Nell and Jelena Cerar – explored methodological trends in IB research, with a particular focus on the data that IB scholars use. It provided an opportunity for the panellists to discuss the prevalence of patterns, their changes over time, and suggestions for future research. The panellists were Bo Nielsen, Catherine Welch, Rebecca Piekkari, Sebastian Reiche and Jelena Cerar, while the panel was chaired by Phillip Nell.

Are primary data dead?

Catherine introduced the Nielsen et al. (2020) paper showing the percentage of JIBS papers published in the last 50 years by method and decade. She outlined the growing scarcity of primary data of any kind – qualitative, survey, mixed methods – and the increasing share of secondary data.

Bo explained why these trends are worrying – and likely to continue, given the rise of ‘Big Data’ and the availability of new databases. He emphasised that all data sources have limitations. Some of the concerns for secondary data include that:

• we do not know how secondary data was collected,
• it was usually collected for another purpose and possibly within a different context, and
• authors may be tempted to engage in data mining.

The most worrying consequence of these trends is that IB might become a more data driven rather than theory driven.

Catherine raised some potential solutions to these problems, such as interrogating the quality of data, as well as data triangulation within studies and within research programs. She emphasised that we need to rethink our priorities regarding data quality vs. quantity, both as reviewers and authors. As reviewers we must be aware that more is not necessarily better (and this applies both to qualitative and quantitative research) and as authors we should explicitly discuss and assess data quality in our methods section.

Why are primary data declining?

Our commentary on the Nielsen et al., 2020 paper (Cerar, Nell and Reiche (2021) shows the declining share of primary data and the neglect of the individual level in the top six IB journals (IBR, JIBS, JIM, JWB, MIR, GSJ). Over the last 20 years, the total cumulative increase for secondary data was 52.6%, while the declining share of primary data (defined as raw data) for the same period was 21.7%. This percentage was even higher (over 30%) when the primary data was defined as data collected for the particular purpose of the published paper. We elaborated on two mechanisms that may explain the observed shifts.

(1) increases in the quality and versatility of secondary data relative to primary data, such as declining response rates and resultant bias in samples and estimates, methodological challenges such as common method variance, limited reproducibility and replicability, and the difficulty to perform longitudinal analyses over long time frames.

(2) the perceived increase in the quality and versatility of secondary data due to higher accessibility and availability of secondary data and its institutional reinforcement over time.

The data across the six journals also demonstrated that secondary data-based research at the individual level remains scarce. This may lead to considerable problems for the field, especially if there is a systematic shift in researchers’ focus from the research question to the data available, risks of reducing theoretical pluralism, and a lack of development of the individual-level mechanisms underlying many IB phenomena.

What are the limitations of interview data?

Rebecca presented the trends and future agenda for the use of qualitative data in IB (Mees-Buss, Welch & Piekkari, 2020). The use of interviews still dominates qualitative IB research (Piekkari, Welch & Paavilainen, 2009). On the other hand, the use of other qualitative data sources - such as annual reports, contracts, news articles, and observations - is limited (Piekkari et al., 2009).

The problem with (mis)using interviews is that most researchers treat interview data as “facts” rather than something that could be problematised. Authors generally assume that interviewees can explain their thoughts and actions. They also mirror what the participants said (i.e., parroting back interviewees’ terms to an academic audience) and expect theoretical concepts to emerge from the interview data.

However, there are limits of interview data. Interviewees may mislead the researcher, and interviewees in IB often speak a different language from the researcher, thereby bringing an additional level of difficulty in interpretation. The best solution to these problems is a healthy scepticism towards interview data. Therefore, the authors (but also the reviewers) need to consider the following questions (Mees-Buss et al., 2020):

• Have the risks of misinterpretation been considered?
• Have the researchers examined their own ‘perceptual screens’?
• Have the interviewees’ social, cultural, and historical contexts been critically understood?
• Have the anomalies (exceptions, negative cases, and minority views) in the data been explained?
• Have rival explanations been investigated?

Discussion and wrap-up

What are the trends in using qualitative data in IB?

Rebecca: Qualitative data are in decline mainly because companies are becoming more reserved toward giving researchers access. Moreover, the more critical qualitative manuscripts are unfortunately not published in JIBS, but rather in Management and Organisation journals.

Given the quality issues Isn’t a good thing that we have less qualitative data in JIBS?

Rebecca: We would miss a lot if we only did research based on secondary data, but we need to complement the interviews with other types of data. It is time to challenge the limitations inherent to both qualitative and quantitative data.

What are secondary data anyway?

Phillip presented the audience with a poll asking which of the listed studies in their opinion use secondary data:

1. a survey of manager perceptions on digital transformation effects,
2. a historical study of trade patterns based on diaries of a merchant during the late Middle Ages,
3. a quantitative meta-analysis of internationalization research,
4. a quantitative study of board announcement texts and their relationship to strategic decision making, etc.

Some examples (e.g., the second one) divided the audience as approximately half of the audience considered the examples as primary data and another half as secondary data. This clearly illustrates that whether we consider data as primary or secondary depends on whether we make a distinction based on the rawness or the original purpose of collected data.

Is GDPR compliance making individual-level data collection even harder?

The panellists generally agreed with this. Rebecca commented that GDPR is becoming more and more of an issue in qualitative research. In the past masters students might have had no problem in doing interviews, but increasingly we need a lawyer instructing us what is and isn’t allowed. This element of formalization and standardization did not exist in the past, at least in Finland. Sebastian added that the introduction of IRB boards (ethics approval) might affect this trend as well. Jelena explained that privacy issues have been relevant for a long time in experimental research, but experimental research has not been used as much in IB as in other disciplines, so we may not have been used to these issues in the past, compared to other disciplines.

References

Cerar, J., Nell, P. C., & Reiche, B. S. (2021). The declining share of primary data and the neglect of the individual level in international business research. Journal of International Business Studies, 1-10.

Mees-Buss, J., Welch, C., & Piekkari, R. (2020). From templates to heuristics: how and why to move beyond the Gioia methodology. Organizational Research Methods, 1-25.

Nielsen, B. B., Welch, C., Chidlow, A., Miller, S. R., Aguzzoli, R., Gardner, E., ... & Pegoraro, D. (2020). Fifty years of methodological trends in JIBS: Why future IB research needs more triangulation. Journal of International Business Studies, 51(9), 1478-1499.

Piekkari, R., Welch, C., & Paavilainen, E. (2009). The case study as disciplinary convention: Evidence from international business journals. Organizational Research Methods, 12(3), 567-589.

Welch, C., Marschan-Piekkari, R., Penttinen, H., & Tahvanainen, M. (2002). Corporate elites as informants in qualitative international business research. International Business Review, 11(5), 611-628.

Related videos

Related blogposts

• The ins and outs of experimental research in IB and Management
• Challenges in International survey research: illustrations and solutions
• Language effects in international mail surveys
• What if fully agree doesn't mean the same thing across cultures?
• AIB RM-SIG: Research methods in International Business
• CYGNA: Secondary data sources and research portfolios
• Experimental research in international management

conferences videos research focus research methods academic publishing international business international research publications

Reddit Twitter LinkedIn Facebook Email