Large Volumes Of Qualitative Data Research Articles

Developing an Analytical Pipeline to Classify Patient Safety Event Reports Using Optimized Predictive Algorithms.

Patient safety event reports provide valuable insight into systemic safety issues but deriving insights from these reports requires computational tools to efficiently parse through large volumes of qualitative data. Natural language processing (NLP) combined with predictive learning provides an automated approach to evaluating these data and supporting the work of patient safety analysts. The objective of this study was to use NLP and machine learning techniques to develop a generalizable, scalable, and reliable approach to classifying event reports for the purpose of driving improvements in the safety and quality of patient care. Datasets for 14 different labels (themes) were vectorized using a bag-of-words, tf-idf, or document embeddings approach and then applied to a series of classification algorithms via a hyperparameter grid search to derive an optimized model. Reports were also analyzed for terms strongly associated with each theme using an adjusted F-score calculation. F1 score for each optimized model ranged from 0.951 ("Fall") to 0.544 ("Environment"). The bag-of-words approach proved optimal for 12 of 14 labels, and the naïve Bayes algorithm performed best for nine labels. Linear support vector machine was demonstrated as optimal for three labels and XGBoost for four of the 14 labels. Labels with more distinctly associated terms performed better than less distinct themes, as shown by a Pearson's correlation coefficient of 0.634. We were able to demonstrate an analytical pipeline that broadly applies NLP and predictive modeling to categorize patient safety reports from multiple facilities. This pipeline allows analysts to more rapidly identify and structure information contained in patient safety data, which can enhance the evaluation and the use of this information over time.

A Method for Developing Trustworthiness and Preserving Richness of Qualitative Data During Team-Based Analysis of Large Data Sets

This article outlines a three-phase, team-based approach used to analyze qualitative data from a nation-wide needs assessment of access to Veteran Health Administration services for rural-dwelling veterans. The method described here was used to develop the trustworthiness of findings from analysis of a large qualitative data set, without the use of analytic software. In Phase 1, we used templates to summarize content from 205 individual semistructured interviews. During Phase 2, a matrix display was constructed for each of 10 project sites to synthesize and display template content by participant, domain, and category. In the final phase, the summary tabulation technique was developed by a member of our team to facilitate trustworthy observations regarding patterns and variation in the large volume of qualitative data produced by the interviews. This accessible and efficient team-based strategy was feasible within the constraints of our project while preserving the richness of qualitative data.

A flipped classroom model for inquiry-based learning in primary education context

A multi-case study will be presented in this publication which aimed to address an important gap in the current literature concerning the effective implementation of a flipped classroom (FC) model in a particular educational setting. There has been limited research focusing on utilising a FC model within the primary education context despite its potential benefits for young students, such as facilitating student-centred inquiry-based learning (IBL) and developing their higher order cognitive skills. This multi-case study has been drawn from authors’ collaborative action research project with other teacher participants, during which the authors explored the effective ways in which a FC model can be utilised to promote students’ IBL in primary school settings. The authors first develop an inquiry-based flipped classroom (IB-FC) model and applied the model into five primary schools in Cyprus for a school year (2017–2018). A total number of five teachers, 77 students and 48 of their parents were invited to participate in the project. A large volume of qualitative data was collected mainly through classroom observations and interviews. Data analysis of teachers’, students’ and parents’ experiences and perceptions led to the development of seven universal design principles. These principles can be used to support primary school teachers’ attempts to design effective instructions using the IB-FC model.

THE METHODOLOGICAL RELEVANCE OF MASS OBSERVATION DATA

The Mass Observation Project, established in 1937, documents the lives of ordinary people living in the UK, and explores a wide range of social issues. The Project distributes a set of written questions (“Directives”) to a panel of 500 members of the British public (“Observers”) three times each year; “Observers” respond in writing. From the initial commissioning of a “Directive” to data becoming available for analysis takes between four to six months. This approach offers researchers an opportunity to capture in-depth qualitative data from individuals with a range of demographic backgrounds who live across the UK. As there are no word limits on “Observers’” responses and they remain anonymous, a “Directive” often yields rich, high-quality data. Additionally, compared with alternative methods of collecting large volumes of qualitative data from a heterogeneous population, commissioning a “Directive” is cost-effective in terms of time and resource.

Taken Out of Context: Hazards in the Interpretation of Written Assessment Comments.

Written comments are increasingly valued for assessment; however, a culture of politeness and the conflation of assessment with feedback lead to ambiguity. Interpretation requires reading between the lines, which is untenable with large volumes of qualitative data. For computer analytics to help with interpreting comments, the factors influencing interpretation must be understood. Using constructivist grounded theory, the authors interviewed 17 experienced internal medicine faculty at 4 institutions between March and July, 2017, asking them to interpret and comment on 2 sets of words: those that might be viewed as "red flags" (e.g., good, improving) and those that might be viewed as signaling feedback (e.g., should, try). Analysis focused on how participants ascribed meaning to words. Participants struggled to attach meaning to words presented acontextually. Four aspects of context were deemed necessary for interpretation: (1) the writer; (2) the intended and potential audiences; (3) the intended purpose(s) for the comments, including assessment, feedback, and the creation of a permanent record; and (4) the culture, including norms around assessment language. These contextual factors are not always apparent; readers must balance the inevitable need to interpret others' language with the potential hazards of second-guessing intent. Comments are written for a variety of intended purposes and audiences, sometimes simultaneously; this reality creates dilemmas for faculty attempting to interpret these comments, with or without computer assistance. Attention to context is essential to reduce interpretive uncertainty and ensure that written comments can achieve their potential to enhance both assessment and feedback.

Application of data analytics for product design: Sentiment analysis of online product reviews

Advanced data analytics is one of the most revolutionary technological developments in the 21st century, which enables the discovery of underlining trends via sophisticated computational methods On various e-commerce and social platforms, millions of online product reviews are published by customers, which can potentially provide designers with invaluable insights into product design. This paper presents a design framework to analyze online product reviews. The objective is to use this machine-generated data to identify a series of customer needs. The framework aims to distill large volumes of qualitative data into quantitative insights on product features, so that designers can make more informed decisions. The framework combines the elements of online product reviews, design theory and methodology, and data analytics to reveal new insights. The effectiveness of the proposal framework is validated through a case study on product reviews from the e-commerce website, Amazon. The framework demonstrates a statistical approach for analyzing online product reviews. The framework acts as an interface between quantitative outputs and the qualitative and creative process of design. Further analysis of results identifies many of incorporating logical, computational methods into the highly subjective and creative process of design.

Flood vulnerability and relocation readiness in Zimbabwe

PurposeThe purpose of this paper is to examine vulnerability conditions to flood disasters in Tokwe-Mukorsi community, Zimbabwe and identifies the barriers that constrained the community from relocating to safe sites.Design/methodology/approachUsing a questionnaire survey, field observations and interviews, the paper examines the biophysical and social vulnerability of the Tokwe-Mukorsi community, Zimbabwe, as well as the barriers that prevented it from relocating to safe sites. A thematic analysis of the large volumes of qualitative data from interviews and walk-through analyses was conducted. Descriptive statistics were used in analysing quantitative data from questionnaires.FindingsResults reveal that households living upstream and downstream of the dam were highly vulnerable to floods. Their biophysical vulnerability was partly induced by the construction of the dam whose basin encroached into the farming and settlement area. The extremely vulnerable group were households living below level of 660 m where five saddle dams had been constructed. The built environment of the community exhibited ignorance of standard building codes. The poor socio-economic conditions of the community and the incessant rains experienced over two weeks contributed to the flood disaster in early 2014. The Tokwe-Mukorsi community failed to relocate to safe places partly due to lack of compensation, absence of basic infrastructure and drought occurrence in relocation sites.Research limitations/implicationsThe calculation of social vulnerability indices is beyond the scope of this study due to non-availability of quantitative data at community level.Originality/valueThis paper provides a comprehensive understanding of why some communities may fail to relocate despite being highly vulnerable to flood disasters.

Farm tourism experiences in travel reviews: A cross-comparison of three alternative methods for data analysis

This paper demonstrates three alternative approaches for mining consumer sentiment from large amounts of qualitative data found in online travel reviews. Manual content coding, corpus-based semantic analysis, and stance-shift analysis represent methods varying greatly in both process and output. For illustration purposes, they are applied in an exploratory study focused on consumers' reaction to farm stays in order to demonstrate how large volumes of qualitative data can be analyzed quantitatively in a relatively efficient and reliable way. A total of 800 narratives describing farm stay experiences and representing four national settings (Australia, Italy, UK, and USA) was collected. The results reveal that each method provides unique insights of what a farm stay vacation evokes, helpful to farm entrepreneurs wishing to develop a tourism business. The findings indicate universal values as recurrent key drivers of customer satisfaction closely relate to rural experiences. Comparing the national datasets, local differences are evident, highlighting regional variations in terms of service products and consumer preferences. From a methodological viewpoint, all three methods produce reliable results by evidence of the similarities across all three analyses. The current study shows manual content coding, corpus-based semantic method, and stance-shift analysis can capture the peculiarities of rural experiences in different national settings.

Evaluate experiences: a qualitative technique to complement quantitative impact assessments

Impact evaluations, whether of aid, development interventions or policy interventions, help hold donors and policy makers accountable for promises made and provide a means to legitimise recommendations for future interventions. Quantitative approaches are the order of the day but do not fully capture the complex and nuanced characteristics of communities important to understand the full effects of interventions and/or identify required interventions. This article describes a cost-effective approach to capturing large volumes of qualitative data in anecdotal form that can help inform interventions and approaches to research. Data collected by using the technique in a poor, urban, mixed race settlement in South Africa are reviewed and used to provide recommendations on research in poverty. For reasons consistent with the theoretical framework explained, recommendations are limited to three: consider carefully intergenerational dynamics when researching poor areas; be responsible about differences in value scales; and be conscious of how differently the researched see the world. A discussion using the method in mixed methods research is provided.

Painting With and Without Numbers: the use of qualitative and quantitative methods to study social learning

The Enterprise Social Learning Architecture (ESLA) team of the Defence Science and Technology Organisation (DSTO), conducted a four-year research study investigating social learning within the Australian Defence Organisation (ADO). The immediate aim of this research was to understand the issues inherent in building learning, adaptive and sustainable systems. The long-term objective was to develop architectures that would support the development of information systems to guide and enhance organisational learning and facilitate knowledge management. In this paper we will discuss the methodologies used by the ESLA team to gain understanding into effective social learning and the organisational and cultural factors that support such learning. Also, the paper will discuss the lessons learned from methodological approaches to this study as well as support tools used to analyse large volumes of qualitative data. There has been an increasing emphasis in the past decade on investigating the social and organisational factors that may underpin successful information system development and usage (Butterfield and Pendegraft, 1996; Davenport and Prusak, 1992; DeLone and McLean, 1992). Investigation of these issues necessitates a sound understanding of organisational culture, human social interactions, communication and relationships, and reflects an increasing awareness of the importance of the social aspects of socio-technical systems that people work and operate in. This paper describes the process by which the qualitative methods in this study of knowledge processes were expanded to include quantitative methods. It focuses on how this combination of data collection methods evolved, and the ways in which it was capitalised on to provide a much more enriched set of findings than would have been the case if qualitative or quantitative methods had been used alone. The paper also focuses on pitfalls that arose in the use of the various methods, including those associated with the use of software for qualitative data, and the means by which a successful methodological blend can be achieved with research relating to knowledge and its management.