Interactive content analysis : evaluating interactive variants of non-negative Matrix Factorisation and Latent Dirichlet Allocation as qualitative content analysis aids

This thesis addressed issues that have prevented qualitative researchers from using thematic discovery algorithms. The central hypothesis evaluated whether allowing qualitative researchers to interact with thematic discovery algorithms and incorporate domain knowledge improved their ability to address research questions and trust the derived themes. Non-negative Matrix Factorisation and Latent Dirichlet Allocation find latent themes within document collections but these algorithms are rarely used, because qualitative researchers do not trust and cannot interact with the themes that are automatically generated. The research determined the types of interactivity that qualitative researchers require and then evaluated interactive algorithms that matched these requirements. Theoretical contributions included the articulation of design guidelines for interactive thematic discovery algorithms, the development of an Evaluation Model and a Conceptual Framework for Interactive Content Analysis.

Driver sleepiness on YouTube: A content analysis

Driver sleepiness is a major contributor to severe crashes and fatalities on our roads. Many people continue to drive despite being aware of feeling tired. Prevention relies heavily on education campaigns as it is difficult to police driver sleepiness. The video sharing social media site YouTube is extremely popular, particularly with at risk driver demographics. Content and popularity of uploaded videos can provide insight into the quality of publicly accessible driver sleepiness information. The purpose of this research was to answer two questions; firstly, how prevalent are driver sleepiness videos on YouTube? And secondly, what are the general characteristics of driver sleepiness videos in terms of (a) outlook on driver sleepiness, (b) tone, (c) countermeasures to driver sleepiness, and, (d) driver demographics. Using a keywords search, 442 relevant videos were found from a five year period (2nd December 2009–2nd December 2014). Tone, outlook, and countermeasure use were thematically coded. Driver demographic and video popularity data also were recorded. The majority of videos portrayed driver sleepiness as dangerous. However, videos that had an outlook towards driver sleepiness being amusing were viewed more often and had more mean per video comments and likes. Humorous videos regardless of outlook, were most popular. Most information regarding countermeasures to deal with driver sleepiness was accurate. Worryingly, 39.8% of videos with countermeasure information contained some kind of ineffective countermeasure. The use of humour to convey messages about the dangers of driver sleepiness may be a useful approach in educational interventions.

Towards automated content analysis of discussion transcripts: A cognitive presence case

In this paper, we present the results of an exploratory study that examined the problem of automating content analysis of student online discussion transcripts. We looked at the problem of coding discussion transcripts for the levels of cognitive presence, one of the three main constructs in the Community of Inquiry (CoI) model of distance education. Using Coh-Metrix and LIWC features, together with a set of custom features developed to capture discussion context, we developed a random forest classification system that achieved 70.3% classification accuracy and 0.63 Cohen's kappa, which is significantly higher than values reported in the previous studies. Besides improvement in classification accuracy, the developed system is also less sensitive to overfitting as it uses only 205 classification features, which is around 100 times less features than in similar systems based on bag-of-words features. We also provide an overview of the classification features most indicative of the different phases of cognitive presence that gives an additional insights into the nature of cognitive presence learning cycle. Overall, our results show great potential of the proposed approach, with an added benefit of providing further characterization of the cognitive presence coding scheme.

Estimating the encounter rate of Atlantic capelin (Mallotus villosus) with fish eggs, based on stomach content analysis

The number of pelagic fish eggs (cod and cunner) found in stomachs of capelin (Mallotus villosus) sampled in coastal Newfoundland was used to estimate the encounter rates between capelin and prey, and thus the effective volume swept by capelin. Fish eggs were found in 4−8% of capelin stomachs, represented an average of 1% of prey by numbers, and their abundance increased as relative stomach fullness decreased. The average number of eggs per stomach doubled for each 5-cm increase in length of capelin. The effective volume swept for eggs by capelin ranged from 0.04 to 0.84 m3/h—a rate that implies either very slow capelin swimming speeds (<1 cm/s) or that fish eggs are not strongly selected as prey. The predation rate estimated from stomach contents was higher than that predicted from laboratory studies of feeding pelagic fish and lower than that predicted by a simple foraging model. It remains uncertain whether capelin play an important regulatory role in the dynamics of early life stages of other fish.

Growth and food availability of silver and bighead carps: evidence from stable isotope and gut content analysis

A 2-year investigation of growth and food availability of silver carp and bighead was carried out using stable isotope and gut content analysis in a large pen in Meiliang Bay of Lake Taihu, China. Both silver carp and bighead exhibited significantly higher delta 13C in 2005 than in 2004, which can probably be attributed to two factors: (i) the difference between isotopic compositions at the base of the pelagic food web and (ii) the difference between the compositions of prey items and stable isotopes. The significantly positive correlations between body length, body weight and stable isotope ratios indicated that isotopic changes in silver carp and bighead resulted from the accumulation of biomass concomitant with rapid growth. Because of the drastic decrease in zooplankton in the diet in 2005, silver carp and bighead grew faster in 2004 than in 2005. Bighead carp showed a lower trophic level than silver carp in 2005 as indicated by stable nitrogen isotope ratios, which was possibly explained by the interspecific difference between the prey species and the food quality of silver carp and bighead.

“It’s just like passing notes in class…”: a content analysis of the use of Twitter at #asl2015

Twitter has changed the dynamic of the academic conference. Before Twitter, delegate participation was primarily dependent on attendance and feedback was limited to post-event survey. With Twitter, delegates have become active participants. They pass comment, share reactions and critique presentations, all the while generating a running commentary. This study examines this phenomenon using the Academic & Special Libraries (A&SL) conference 2015 (hashtag #asl2015) as a case study. A post-conference survey was undertaken asking delegates how and why they used Twitter at #asl2015. A content and conceptual analysis of tweets was conducted using Topsy and Storify. This analysis examined how delegates interacted with presentations, which sessions generated most activity on the timeline and the type of content shared. Actual tweet activity and volume per presentation was compared to survey responses. Finally, recommendations on Twitter engagement for conference organisers and presenters are provided.

A content analysis method to measure critical thinking in face-to-face and computer supported group learning

This paper gives a detailed account of the content analysis method developed at Queen's University Belfast to measure critical thinking during group learning, as used in our controlled comparisons between learning in face-to-face and computer conference seminars. From Garrison's 5 stages of critical thinking, and Henri's cognitive skills needed in CMC, we have developed two research instruments: a student questionnaire and this content analysis method. The content analysis relies on identifying, within transcripts, examples of indicators of obviously critical and obviously uncritical thinking, from which several critical thinking ratios can be calculated.

Predatory nature of the littoral amphipod Echinogammarus marinus: gut content analysis and effects of alternative food and substrate heterogeneity

Interspecific interactions are major structuring forces in marine littoral communities; however, it is unclear which of these interactions are exhibited by many key-component species. Gut content analysis showed that the ubiquitous rocky/cobble shore amphipod Echinogammarus marinas, often ascribed as a mesograzer, consumes both algae and macroinvertebrates. Further, laboratory experiments showed that E. marinus is an active predator of such macroinvertebrates, killing and consuming the isopod Jaera nordmanni and the oligochaete Tubificoides benedii. Predatory impacts of E. marinus were not alleviated by the presence of alternative food in the form of alga discs. However, in the presence of prey, consumption of alga by E. marinus was significantly reduced. Further, survival of prey was significantly higher when substrate was provided, but predation remained significant and did not decline with further increases in substrate heterogeneity. We conclude that such amphipods can have pervasive predatory impacts on a range of species, with implications for community structure, diversity and functioning.

What the newspapers say about medication adherence: a content analysis

Background: This study investigates the coverage of adherence to medicine by the UK and US newsprint media. Adherence to medicine is recognised as an important issue facing healthcare professionals and the newsprint media is a key source of health information, however, little is known about newspaper coverage of medication adherence.

Methods. A search of the newspaper database Nexis®UK from 2004-2011 was performed. Content analysis of newspaper articles which referenced medication adherence from the twelve highest circulating UK and US daily newspapers and their Sunday equivalents was carried out. A second researcher coded a 15% sample of newspaper articles to establish the inter-rater reliability of coding.

Results: Searches of newspaper coverage of medication adherence in the UK and US yielded 181 relevant articles for each country. There was a large increase in the number of scientific articles on medication adherence in PubMed® over the study period, however, this was not reflected in the frequency of newspaper articles published on medication adherence. UK newspaper articles were significantly more likely to report the benefits of adherence (p = 0.005), whereas US newspaper articles were significantly more likely to report adherence issues in the elderly population (p = 0.004) and adherence associated with diseases of the central nervous system (p = 0.046). The most commonly reported barriers to adherence were patient factors e.g. poor memory, beliefs and age, whereas, the most commonly reported facilitators to adherence were medication factors including simplified regimens, shorter treatment duration and combination tablets. HIV/AIDS was the single most frequently cited disease (reported in 20% of newspaper articles). Poor quality reporting of medication adherence was identified in 62% of newspaper articles.

Conclusion: Adherence is not well covered in the newspaper media despite a significant presence in the medical literature. The mass media have the potential to help educate and shape the public's knowledge regarding the importance of medication adherence; this potential is not being realised at present. © 2013 Goodfellow et al.; licensee BioMed Central Ltd.

High content analysis: A sensitive tool to detect and quantify the cytotoxic, synergistic and antagonistic effects of chemical contaminants in foods

Aflatoxin B1 (AFB1), ochratoxin A (OTA) and fumonisin B1 (FB1) are important mycotoxins in terms of
human exposure via food, their toxicity and regulatory limits that exist worldwide. Mixtures of toxins can frequently be present in foods, however due to the complications of determining their combined toxicity,
legal limits of exposure are determined for single compounds, based on long standing toxicological
techniques. High content analysis (HCA) may be a useful tool to determine total toxicity of complex
mixtures of mycotoxins. Endpoints including cell number (CN), nuclear intensity (NI), nuclear area (NA),
plasma membrane permeability (PMP), mitochondrial membrane potential (MMP) and mitochondrial
mass (MM) were compared to the conventional 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium
bromide (MTT) and neutral red (NR) endpoints in MDBK cells. Individual concentrations of each
mycotoxin (OTA 3mg/ml, FB1 8mg/ml and AFB11.28mg/ml) revealed no cytotoxicity with MTTor NR but
HCA showed significant cytotoxic effects up to 41.6% (p0.001) and 10.1% (p0.05) for OTA and AFB1,
respectively. The tertiary mixture (OTA 3mg/ml, FB1 8mg/ml and AFB1 1.28mg/ml) detected up to 37.3%
and 49.8% more cytotoxicity using HCA over MTT and NR, respectively. Whilst binary combinations of
OTA (3mg/ml) and FB1 (8mg/ml) revealed synergistic interactions using HCA (MMP, MM, NI endpoints)
not detected using MTT or NR. HCA is a highly novel and sensitive tool that could substantially help
determine future regulatory limits, for single and combined toxins present in food, ensuring legislation is based on true risks to human health exposure.

Online health information - what the newspapers tell their readers: a systematic content analysis

Background: This study investigated the nature of newspaper reporting about online health information in the UK and US. Internet users frequently search for health information online, although the accuracy of the information retrieved varies greatly and can be misleading. Newspapers have the potential to influence public health behaviours, but information has been lacking in relation to how newspapers portray online health information to their readers.

Methods: The newspaper database Nexis (R) UK was searched for articles published from 2003 - 2012 relating to online health information. Systematic content analysis of articles published in the highest circulation newspapers in the UK and US was performed. A second researcher coded a 10% sample to establish inter-rater reliability of coding.

Results: In total, 161 newspaper articles were included in the analysis. Publication was most frequent in 2003, 2008 and 2009, which coincided with global threats to public health. UK broadsheet newspapers were significantly more likely to cover online health information than UK tabloid newspapers (p = 0.04) and only one article was identified in US tabloid newspapers. Articles most frequently appeared in health sections. Among the 79 articles that linked online health information to specific diseases or health topics, diabetes was the most frequently mentioned disease, cancer the commonest group of diseases and sexual health the most frequent health topic. Articles portrayed benefits of obtaining online health information more frequently than risks. Quotations from health professionals portrayed mixed opinions regarding public access to online health information. 108 (67.1%) articles directed readers to specific health-related web sites. 135 (83.9%) articles were rated as having balanced judgement and 76 (47.2%) were judged as having excellent quality reporting. No difference was found in the quality of reporting between UK and US articles.

Conclusions: Newspaper coverage of online health information was low during the 10-year period 2003 to 2012. Journalists tended to emphasise the benefits and understate the risks of online health information and the quality of reporting varied considerably. Newspapers directed readers to sources of online health information during global epidemics although, as most articles appeared in the health sections of broadsheet newspapers, coverage was limited to a relatively small readership.