Skip to content

H809 – Reading Wegerif and Mercer (1997) A2.4a

February 13, 2011

Here are my answers to the basic questions we were introduced at the beginning of H809.

  1. Questions: What research questions are being addressed?
  2. Settings: What is the sector and setting?
  3. Concepts: What theories, concepts and key terms are being used?
  4. Methods: What methods of data collection and analysis are used?
  5. Findings: What did this research find out?
  6. Limitations: What are the limitations of the methods used?
  7. Ethics: Are there any ethical issues associated with the research?
  8. Implications: What are the implications (if any) for practice, policy or further research?

Research question(s)

Can computer-based analysis of collaborative learning effectively combine qualitative and quantitative discourse analysis method to reduce the weaknesses of both methods?

Does a change in language of the children towards exploratory talk lead to an increase in group reasoning test scores?

Setting

Primary school classrooms

Concepts

I am not sure about theories and concepts used in the study, besides that they explored collaborative learning and how the use of language of the children changed towards exploratory talk in the post-intervention task while doing reasoning tests.

However, a couple of key terms are used like computer-based text analysis, qualitative and quantitative research methods, level of data and abstraction, key word in context analysis and count of key word usage.

Methods

The computer-based analysis of collaborative learning combines qualitative methods of discourse analysis with quantitative methods. The authors used !KwicTex, a concordancer designed by Graddol, which allows the researcher to move between full transcripts and lists of utterance, or other contextual unit, containing key words in context.

The intervention programme consisted of a series of eight lessons coaching exploratory talk and each lesson integrated three stages, starting with an explanation given by the teacher, followed by a small group exercises and finished with a whole class discussion with reflection.

The participating children all aged 9 an 10 worked together in groups of three. In the target class, nine groups produced comparable pre- and post-intervention test, contrary to five groups in the control group.

Three ‘focal groups’ in the target group were video-recorded to better look at changes in quality of the interactions of the groups.

Findings

Wegerif and Mercer, 1997 found out that computer-based analysis of collaborative learning effectively combine qualitative and quantitative discourse analysis method and thus reduce the weaknesses of both methods.

Computer-based transcripts analysis of collaborative learning ‘can produce an overall interpretation which is more convincing than either qualitative or quantitative accounts can be if used alone. Computer-based methods of text analysis generate data that allows ‘researcher to move rapidly between different levels of abstraction in dealing with transcript data.

Within the study Wegerif and Mercer conducted they reached two conclusions.

  1. the exploratory features of talk responsible for solving problems were features generally found more in the post-intervention talk of the children doing a reasoning test task than in the pre-intervention talk.
    Qualitative analysis using the ‘key word in context’ or count of ‘key usage’ analysis (see transcripts and table 1) facilitated this generation of data.
  2. the change in the style of talk of the children towards exploratory talk was matched with an increase in group reasoning test scores (Quantitative method, see figure 1).

Hence, there is a positive relationship between the change in the use of language and the improvement in test-scores. The more the children apply exploratory talk the better their test results.

Limitations

Computer-based analysis sees a pretty good method to combine qualitative and quantitative methods. The advantages of each method is used while minimizing respective disadvantages, thus leading to greater validity and reliability of research data. However, I almost assume, though I don’t see it right now, that this method might have a hitch as well. Probably the software !KwicTex is too expensive or it is too difficult to use. 😕

Ethics

I found this ethical considerations in my K221 course book:

  • the need for ethical approval
    – obtain ethical approval from an appropriate body
    – the interest of people taking part are fully protected, they gave their informed consent to taking part
  • Bias and credibility
    – People who conduct the evaluation should not be too intimately involved in the project to avoid bias
    – Researchers should have credibility with the various stakeholders
  • Disclosure and potential conflicts of interest
  • Honesty and competence
  • Ensure the protection of people taking part
    confidentiality and anonymity for participating people, ensuring that they have given their informed consent
  • systematic, accurate and fair evaluation
  • show report to interested parties before wide dissemination

I assume that parent’s gave their informed consent to the evaluation of their children’s talk, and although they use the name of the children the location and name of the primary school was not given, thus identification fo the children should be not given.

Implications

Wegerif and Mercer propose that the computer-based method that combines qualitative and quantitative methods ‘produce an overall interpretation which is more convincing than either qualitative or quantitative accounts can be if used alone’. Further research could be thus improved using computer-based analysis, because ‘it enables the abstraction of different levels of linguistic data without ever leaving behind the original linguistic contexts of the actual words spoken. Often research has less to offer for practice because research  has often ignored issues that are important in current practice, e.g. mapping course design, contexts and the types of interactivity. The vague and abstract concepts leaves the practitioner with a wide range of choices how to implement a certain technology and to find ways of achieving the positive outcomes that others have reported. As this approach combines 4 different levels of abstraction ranging from a general level of data reached by the results of the group reasoning tests to the most concrete level of data represented by the full transcript of each recorded event. This helps to bridge  from research to practice and could help to improve learning outcomes.


That was a hard piece of work and I am pretty sure that after finishing my degree I am not planning to become a researcher. I recently had a really good telephone interview with an OU career adviser to shed some light into my possible career options and we figured out that my main interests are in developing elearning activities, thus an elearning developer. But I also know that keeping up-to-date with research, to critically evaluate and reflect on it, is in the fast changing field of elearning is a prerequisite for all professionals working in the field of elearning.

Coming back to the Wegerif and Mercer’s paper. Computer-based text analysis was new to me and I found the potential it offers to research speech interesting. But the reflection on th paper will be subject of the next activity and blog post.

One issue that confused me was the interpretation of Figure 1. Looking at Figure 1 I see a greater increase in the control group as in the target group between pre-test and post test. The controls start with a lower mean score (15,5) but increased by five point up to 20,5. The targets, although the had a mean score of almost 17 in the pre-test, they only reached a result of just above 19, thus only an increase from about 3 points. However, the authors offer a different interpretation. They claim that the target class group scores increased by 32%, contrary to the control class group scores which increased by 15%. Can anybody please enlighten me? I know that nine groups produced comparable pre- and post intervention test in the target class and only five groups in the control group. But does that explain our different interpretations? I know that different sample groups can confound the results, but can they explain the differences between Figure 1 and the results given in the paper?!

References

Wegerif, R. and Mercer, N. (1997) ‘Using computer-based text analysis to integrate qualitative and quantitative methods in research on collaborative learning’, Language and Education, vol. 11, no. 4, pp. 271–86.

Open University (2005) K221 Complementary and Alternative Medicine: Structures and Safeguards, Chapter 9 ‘Evaluating CAM practice’ in Le-Treweek, G., Heller, T., MacQueen H., Stone, J. and Spurr, S., Milton Park, Routledge.

Advertisements

From → H809

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: