Text Mining in Organizational Research

text-mining-1476780_1920[We’re pleased to welcome authors Dr. Stefan Mol, Vladimer B. Kobayashi, Hannah A. Berkers, Gabor Kismihok, and Deanne N. Den Hartog of the University of Amsterdam. They recently published an article in Organizational Research Methods entitled “Text Mining in Organizational Research,” which is currently free to read for a limited time. Below, Dr. Mol recounts the events that led to the research and the significance it has to the field:]

07ORM13_Covers.inddWere there any specific external events that influence your decision to pursue this research?
One critical on-going event that lead us to pursue this research is the revolution and promise brought by the rise of big data to understand and enhance organizational processes. A large proportion of these data are comprised of texts that are generated every day at rates that imply that manual analysis of all of this data is no longer possible. The abundance of untapped text data suggest the existence of information with the promise of generating new knowledge that may be used to enhance both individual and organizational level outcomes.
Although, organizations already collect and store text data, many do not fully take advantage of the knowledge that can be gleaned from analyzing text. This may be due to a lack of expertise in conducting automatic text analysis or text mining. The mission of our work here is to empower organizational researchers by raising awareness of the possibilities afforded by text mining, helping them see how text mining might help them answer their research questions, and helping them to understand and use the text mining process and tools.
In what ways is your research innovative, and how do you think it will impact the field?
With this article we hope to contribute by facilitating dialogue between data scientists and organizational researchers about the opportunities afforded by text mining. As an example, we illustrate the role that text mining of vacancies might play in job analysis. Previous approaches to job analysis rely on time consuming collection and analysis of survey and observation based data the results of which soon become outdated due to the fast changing nature of jobs. Using text mining we demonstrate how one can take advantage of other data sources such as online job vacancies to understand the requirements and skill demands of different types of jobs. Our goal is to not only apply text mining to the field of job analysis but more importantly to inform organizational researchers about the wide-ranging uses text mining could have in organizational research. We hope that this will spark an increase in the use of text data and machine learning in organizational research.
What advice would you give to new scholars and incoming researchers in this particular field of study?
Existing text mining solutions are technique and tool-oriented because most techniques and Big Data tools are currently primarily shaped by technical fields, such as statistics and computer science, that put greater emphasis on the computational and technological aspects. However, applying these in the field of organizational research holds great promise. Organizational researchers bring with them a repertoire of organizational theories. These theories provide domain specific information and requirements that can influence the selection of techniques and analytical strategy, and the way to evaluate the success of the particular application. Our advice for incoming organizational researchers wanting to explore text mining is to draw on their own theoretical expertise and from there start selecting the appropriate techniques and approaches to text mining. Also, as with using other analytical tools, we do need to pay careful attention to rigor in evaluation and validation of text mining based results.

Stay up-to-date with the latest research from Organizational Research Methods and sign up for email alerts today through the homepage!




Text Classification for Organizational Researchers: A Tutorial

baby-84626_1920[We’re pleased to welcome author Dr. Stefan Mol of the University of Amsterdam. Dr. Mol recently published an article in Organizational Research Methods entitled “Text Classification for Organizational Researchers: A Tutorial,” which is currently free to read for a limited time. Below, Dr. Mol reflects on the inspiration for conducting this research:]

07ORM13_Covers.inddWhat motivated you to pursue this research?
Machine Learning assisted text analysis is still uncommon in organizational research, although its use holds promise. Most manual text analysis procedures conducted by researchers in this field are about the assignment of text to categories such as in thematic and template analyses. However, manual classification of text becomes laborious and time consuming (and sometimes subject to reliability issues) when one needs to do this for a sizeable amount (hundreds of thousands or millions) of pieces of text. An alternative is to use automatic text classification systems that can be constructed by researchers, which allow them to speed up the process of labeling or coding large sets of textual data. The design and building of text classifiers could be of use for various areas of organizational research. Our aim was to illustrate how this could be done and provide a tutorial. We used the example of building a text classifier to automatically sort job type information contained in job vacancies. The importance of validating the results of text classification was demonstrated through data triangulation, using expert input. We believe that the use of this procedure among organizational researchers can improve reliability and efficiency in analysis that involves classification.
What has been the most challenging aspect of conducting your research? Were there any surprising findings?
Building classifiers involves several rounds of training, testing, and validation before they can be deployed in practice and the most challenging aspect is training the classifier and choosing the parameters in such a way that the results are valid from the standpoint of application. The classifier we built for the job analysis task was able to recover job task sentences with high precision as assessed by an expert in the field, although the classifier was initially trained with minimum expert input. Our results thus suggest that job vacancies are a reliable alternative source of job information that can augment existing approaches to job analysis. More generally, we believe this also suggests that wider use of text classification holds promise for organizational research in a broader sense.
What did not make it into your published manuscript that you would like to share with us?
One class of techniques that are now increasingly applied in the area of text classification are word embeddings. Word embeddings map each word to vectors of real numbers. The similarities among word vectors can be used to quantify and categorize the meaning of words in specific contexts. We initially planned to include a short discussion about this but we decided not to because these techniques warrant more in depth discussion which go beyond the scope of our current article. However, organizational researchers interested in recovering context specific meaning of words may benefit from the specific approach taken with word embeddings and we recommend them to get to know these techniques as well.

Stay up-to-date with the latest research from Organizational Research Methods and sign up for email alerts today through the homepage!




Using Text Mining for Customer Feedback

[We’re pleased to welcome Francisco Villarroel Ordenes, who is one of five collaborating authors on the article “Analyzing Customer Experience Feedback Using Text Mining: A Linguistics-Based Approach” from Journal of Service Research.]

The Big Data phenomenon is not only about exponential growth of customer data, but about new and challenging data structures such as textual information which require new methods and metrics to facilitate 02JSR13_Covers.inddanalysis. Customer experience feedback, usually found in platforms such as social media, e-mails and feedback forms represents a form of complicated data structure which is challenging organizations to develop new methods for its timely and consistent analysis. Our paper, “Analyzing Customer Experience Feedback Using Text Mining: A Linguistics-Based Approach”, is the result of a collaborative effort between Marketing and Information Systems researchers. We develop a Case Study with a UK service organization which receives more than 10000 comments of customer experience feedback per month. In this context, we design and implement the ARC (Activities, Resources, Context) framework, which is able to automate the analysis of customer feedback through a text mining model. The text mining approach used with this guiding framework is useful for analyzing customer experience feedback with the standard flow of activities (stages) of any service. Due to its flexible evolutionary format we describe it as an ‘open learning model’. Specifically, application of the text mining model within the ARC framework provides efficient and faster analysis of textual information compared with the current manual processes (seconds compared with 2 weeks). The consistency of the information extracted and the specificity of the analysis provided deliver an additional advantage: namely, the practicality of identifying resources or activities that the company can improve immediately. The article provides managers and researchers with a text analytics methodology and application which departs from simple sentiment analysis. Instead a more holistic representation of customer experience feedback in verbatim data is identified, which enables managers to identify what is causing particular sentiment outcomes and thus they can then act to reallocate resources or change processes at an organizational or even customer-specific level.

Read “Analyzing Customer Experience Feedback Using Text Mining: A Linguistics-Based Approach” from Journal of Service Research for free by clicking here. Make sure to click here to sign up for e-alerts and be the first to know all the latest from Journal of Service Research!

s200_francisco.villarroelFrancisco Villarroel Ordenes is a PhD candidate at the Marketing and Supply Chain Management Department at the School of Business Economics in Maastricht University. His research interests include social media conversations, customer experience feedback, sentiment analysis, value cocreation, and the development of text mining methods for marketing research.

publicphoto.ashxBabis Theodoulidis is an associate professor in information management at Manchester Business School, University of Manchester. His research has been published in both science and social science journals such as International Journal Services Technology and Management, Journal of Information Systems, Knowledge Management Research & Practice, Expert Systems with Applications, International Journal of Information Management, International Journal of Data Warehousing and Mining, and Journal of Visual Languages and Computing. His most recent research interests focus on the design of service-based information systems, the temporal and spatial aspects of information, the analysis of information using data and text mining techniques, the visualization of information, and service information management issues within organizations.

jamie.ashxJamie Burton is head of the marketing group and an associate professor in Marketing at Manchester Business School (MBS). He is a research director for MBS’s Customer Management Leadership Group, publishes in a number of journals including the Journal of Marketing Management and the Journal of Service Management and his research interests include customer experience and feedback, transformative service research including service marketing, servitization, relationship marketing, and customer profitability. He is a lead author of a 2013 British Quality Foundation report and is coauthor of Murphy, J. et al. (2006), Converting Customer Value: from Retention to Profit, Chichester: John Wiley and Sons.

Thorsten_GruberThorsten Gruber is a co-director of the Centre for Service Management and a professor of Marketing and Service Management at Loughborough University. His research interests include consumer complaining behavior, services marketing, and the development of qualitative online research methods. His work has been published in journals such as Journal of the Academy of Marketing Science, Journal of Product Innovation Management, Journal of Business Research, Journal of Service Management, and Industrial Marketing Management.

MZMohamed Zaki is a research associate at Cambridge Service Alliance, University of Cambridge. His research lies in the field of information governance, business intelligence, and big data analytics. He has many publications in these areas. His experience in the business intelligence/data analytics and service innovation areas enables him to consult in various projects to investigate business intelligence issues in different domains within a service-oriented architecture. Currently, he investigates “How Big data could play a role in improving and optimising services within complex service network organisation” in different sectors such as education, asset heavy, and defense.