8 Evaluating like a social scientist

There is a curious distinction argued in some of the literature on course evaluation where the concepts of evaluation and research are treated as distinct and disparate activities. It may be because evaluation has often been the function of a separate university unit or a distinction as a result of the teaching/research tension. Whatever the cause, it has meant that the range of analytical research tools brought to bear on evaluation has been somewhat limited. This section opens up the social science researcher's toolkit for inspiration in the evaluation process and argues that evaluation is a form of research, into learning and into the ways in which disciplinary knowledge is constructed and transmitted. Now that academic teachers are more often involved in course evaluation, it would be a shame not to draw on their collective research experience in this context.

One feature shared by both types of inquiry is the often elusive nature of the subject which Bruno Latour might suggest takes place in a black box. Student learning, as we have seen, is typically measured by the assessment process and quantified in student results. It is very difficult to find hard scientifically testable extrinsic evidence of the actual process of learning, although some scientists have used MRI and other scanning technologies to try to gain a better understanding. Typically learning research is done by observing the effects of learning on individuals and groups, the environments in which they learn and the attitudes of participants. Much like other social science research, patterns sought and hypothesis are advanced and tested to explain those patterns.

Much of the data about learning comes from discussion with the learners themselves. A variety of methods, from interviews, to focus groups and surveys are utilised to gather discursive data from which we can infer data about the learning which is happening ‘under the hood‘. Good instrument design is crucial, asking questions that do not lead by suggesting answers and that dig beneath surface knowledge; keeping respondents informed as to the rationale beneath the question. Poorly designed instruments gather little more than surface impressions, a flaw which some anonymous student surveys also suffer.

Other methods break the fourth wall between researcher and the learning environment. Participant observation, action research and auto-ethnography are established social science research methods that can yield deep contextual knowledge that cannot be obtained by ‘white coats and clipboards‘. There is a wealth of literature and online information on the effective design and uses of these research methods but once again it can be much easier just to tap into the expertise already on staff.

Once data is gathered it is essential that you are able to examine and analyse it. Quantitative data requires a particular skillset in statistical analysis and validation, qualitative data has its own frameworks and analysis methods. Here again the evaluation process can useful draw on experts in these fields for guidance in data analysis. One common challenge is ensuring that a sample size is large enough to make valid observations about learning, as students can be reluctant to participate for various reasons. Given a class or course size, a statistician will be able to tell you how large a sample you need to be credible.

Another intersection of research and evaluation is the learning and teaching grant, where recipients are required to evaluate the success of their project on completion. The methods described here, and throughout the rest of this manual should be of value in preparing the evaluation aspect of grant application documents and thinking beyond the simple online survey which is relied on very frequently.

Ethical evaluation and the use of social science methods
Social science researchers have experience in obtaining ethical clearance for research projects involving confidential interviews and other interactions with human subjects. One unfortunate consequence of the artificial evaluation/research distinction is that there can be a lack of experience on evaluation teams and even a lack of awareness regarding the need for ethics approval.

Some evaluation units have been closely tied to the administration of universities and see evaluation as an internal matter that does not require oversight. Whether or not this was true in the past, the new evaluation culture requires a keen awareness of the legal and ethical responsibilities of data gathering particularly where confidential and controversial matters may be under discussion. Students, and other participants, need the protection of ethical oversight and we can draw on the experience of academic researchers, particularly those who have used the same methods in their research previously.

Action research and closing the loop
While social scientists once tried to distance themselves from research subjects in order to don the lab coat of objectivity, action research oriented practice finds objectivity in robust methodological design and emphasises the fully informed participation of subjects. Researchers have acknowledged a responsibility to subjects and to the community at large to share the results and even sometimes to allow subjects to collaborate on the future direction that research takes. In the evaluation context this responsibility also manifests itself through closing the feedback loop.

Effective representation and visualisation of data is an important issue here as learners are unlikely to have the same time to focus on the evaluative aspects of their course, being somewhat focussed on the primary business of learning. Section 11 discusses the effective use of learning analytics and reiterates the importance of the information visualisation process, an increasingly central aspect of the research process.

While learners do not need to be swamped with statistics, tables and charts, the feedback process is a useful juncture for them to learn more about their own learning and the premises upon which their learning environments are being constructed. Students who are better informed are more able to enhance their own learning and to participate in the construction of the learning community. Some of these students will be also be the academics and evaluation managers of the future who will one day be designing evaluation systems of their own.


Methods
  • Participant Observation
  • Structured and semi-structured interviews
  • Focus groups
  • Ethnography and learning portfolios



Key Points


  • Course evaluation does not exist in a vacuum and it should draw on the expertise and innovation of researchers in other fields, especially social scientists who are accustomed to gathering data about people and social environments.
  • Closing the loop is an essential expression of respect for the students and others who take the time to engage with the evaluation process.
  • Look within your organisation for people skilled in qualitative and quantitative data methods, in the analysis and presentation of data and in obtaining ethical clearance.




Participant Observation
In a nutshell:

Rather than seeking to be an impartial outsider, observation is conducted from the inside of a learning activity. This is not so effective in a traditional hierarchical classroom but can be very useful where educational activities are co-produced and teachers are active collaborators in learning.

Example Questions:

How was I, a teacher, involved in the learning of others?

How effectively do I model learning strategies?

Reporting:

This is would generally be a reflective self-assessment process but it may be shared with others as research.

Pros:

Can be a unique point of view to observe learning actually happening.

There is a body of social science research theory on the method, particularly concerned with objectivity and bias.

Cons:

A time consuming process which may be difficult to write up in a way that can be shared and published.

Caution:

As with social research, careful objectivity and impartiality protocols need to be set up and maintained. Particular care must be taken in transitioning to the assessor role so that students do not feel betrayed by someone who they had assumed to have a friendly rather than professional relationship.



Structured/Semi Structured Interviews
In a nutshell:

Interviews conducted on the basis of a set of structured or semi-structured questions, determined in advance and cleared through ethics channels.

Example Questions:

What aspects of this course helped you develop your information technology skills?

Did the new learning app help you to think differently about this area of knowledge?

Who is the most influential person when it comes to the way in which you manage teams?

Reporting:

Generally published as research, but can be part of other reporting mechanisms.

Pros:

Interviews can seek out focussed information that may not be accessible through more formal instruments like surveys.

For semi-structured interviews, the ability to ask further questions and to allow respondents to elaborate.

Plenty written about the method.

Cons:

Time consuming and costly, particularly if interview recordings need to be transcribed (which is not always necessary)

Survey fatigue also applies here.

Caution:

Recruitment can be difficult to do and to provide appropriate incentives for. Incentives need not be financial, but participants should see some direct benefit from being involved.







Focus Groups
In a nutshell:

Focus groups are similar to interview but allow group dynamics to be analysed.

Example Questions:

If we asked you to solve this problem, how would you go about it?

Reporting:

Generally published as research, but can be part of other reporting mechanisms.

Pros:

Ability to examine social aspects of learning.

May be more time effective than interviewing individuals separately.

May enhance collective ownership of the process.

Plenty written about the method.

Cons:

Time and resources

Caution:

This method has been developed by advertising and it if not carefully managed can quickly devolve into surface discussion of personal likes and dislikes rather than the planned questions.





Ethnography & Learning Portfolios
In a nutshell:

The ethnographic methods involves focussed investigation of life stories and narrative in a particular context. In social sciences this method seeks a deep understanding of the personal circumstances of and individual or a community. For learning research ethnography places students in a variety of contextual networks, not simply as a learner in a classroom. Auto-ethnography is an established method which is gaining prominence as we realise that researchers are also located in context with their field of inquiry.

Example Questions:

How does learning interface with other aspects of students' lives?

What does career mean for students in this course?

How do childhood learning experiences impact on adult learning styles?

Reporting:

Generally published as research, but can be part of other reporting mechanisms.

Pros:

An in-depth understand of learning as it relates to specific individuals

Investigates how individuals create meaning out of context

Can bring out structural issues underlying an environment

There is a body of literature on methods.

Cons:

A very labour intensive method which requires extensive resources and demands on participants.

Must be careful in making generalisations based on the experiences of individuals.

Caution:

Student ePortfolios might provide a source for ethnographic research, so long as proper consent and ethical clearance is obtained.

No comments:

Post a Comment