Digital Methods

Digital Assessment

Overview

Generic placeholder image

Controversy mapping

Digital methods used to investigate “public contestation over topical affairs” (Marres, 2015, p. 9) – or in this case, the absence of debate.

Generic placeholder image

Assessment

“What’s intelligence? That’s what my tests measure!” (apocryphal remark attributed to Alfred Binet, inventor of the IQ test

Generic placeholder image

Centres of Calculation

‘sites where literal and not simply metaphorical calculations are made possible by the mathematical or at least arithmetic format of the documents being brought back and forth’ (Latour, 2005, p. 181)

Generic placeholder image

Theory

Actor-network theory and praxiography constitute the principal theoretical resources while trace ethnography and software studies were the methodological approaches used to follow the distributed agency.

Generic placeholder image

Trace ethnography

“Combines the richness of participant-observation with the wealth of data in logs so as to reconstruct patterns and practices of users in distributed sociotechnical systems.” (Giger and Ribes, 2011)

Generic placeholder image

Code

“Code, woven into the background of transactions, habits and perceptions, does not often become visible, except in breakdowns, failures and at certain other moments.” (Mackenzie, 2006, p. 170).

Questions


1: How are the technologies that e-assess adult literacy skills enacted in European educational policy and practice?

2: How are standards made and deployed through local and international e-assessment networks?

3: How are data about people produced through e-assessment events?


The so-called ‘data revolution’ has been transforming many aspects of our lives. Ranging from the ‘microdata’ used to construct portraits of individuals to the aggregated models of big data, the technologies of data production have been assuming an increasingly important role in educational practices. These have been accompanied by new ways of visualizing, sharing and presenting digital data that have in turn have been influencing how educational attainment is represented, managed and, performed.
Much of this is due to novel digital technologies that, in addition to intensifying older data production techniques, allow for data to be produced and distributed in larger quantities and more quickly than ever before. The pervasive and voracious character of digital data production means that it acts as a constant gardener patiently ‘harvesting' data on the ability, behaviour, identity and habits of people through measurement and classification.
As educational practices are becoming increasingly digital, new ways of tracing and understanding the contingencies, hiatuses, compromises, and controversies that come with the folding of social and technical actants into the everyday courses of action that constitute education are necessary.
It describes a methodological approach, trace ethnography, that allows researchers to follow the distributed agency of digital actants such as code.
Sites where digital assessment practices are performed are explored as fruitful areas of investigation. These oligopticons provide narrow but extremely detailed views of individuals and collectives that are slowly built up log-file by log-file, data point by data point.
The presentation concludes with suggestions for how researchers, practitioners and other stakeholders can make decisions informed by an understanding of how data are produced by the various epistemic communities that perform the work of data production.


Controversy map of PIAAC headlines 8-10 October, 2013 using the Actor Network Text Analyser

An uncontroversial controversy

"A Simple Equation: More Education = More Income"

"Americans Think We Have the World’s Best Colleges. We Don’t."

"It’s Now the Canadian Dream"

"The United States, Falling Behind"

"U.S. Adults Fare Poorly in a Study of Skills"

"Stubborn Skills Gap in America’s Work Force"

Standards

"Classifications and standards give advantage or give suffering. Jobs are made or lost; some regions benefit at the expense of others" (Bowker & Star, 1999, p. 6)

Literacy
Competencies and/or Skills
Human Capital
Mulilingual Literacy
Language Skills or Return on Investment (ROI)

PIAAC National Rankings for literacy

Measurable Standards

"The whole enterprise is under the powerful influence of growing public and government demands for tests and measurements that will meet the requirement of accountability and provide for more efficient education without the need to carry out major reforms or to increase resources." (Spolsky, 2008, p. 301)

The post-war project witnessed a convergence of several phenomena: of comparative education. Psychometrics used as a way to create a measurable construct: literacy. This in turn was linked to statistical analyses that were able to produce data for macro-econmics that were able to be translated into skills and human capital.

Sankey Chart



Assessment (or assessment-as-testing)

Strictly speaking, a theory of assessment does not really exist" (Chatel 2008)


Assessment as a tool

Assessment as data /artefact producer

However, assessment can be understood and investigated as a practice

Can assessment be investigated ethnographically?

(yes)

Assessment as a socio-material practice

"An unpredictable and unstable processes of ‘translation’ as various actors and objects come together to form a network that constitutes particular settings and events" (Maddox, 2014, p. 2)

Praxiography - "starts with practices, situations, and events in which information technologies appear, asking openly what occurs and what emerges" (Strand, 2010, p. 5)

Two assessment stories

Centres of calculation

One 'local smal- scale assessment site: Organization Z.

One international large-scale international assessment site: The Organisation for Economic Cooperation and Development (OECD)

Code, Events and Actants

"Code is a set of permutable distributions of agency between people, machines and contemporary symbolic environments carried as code. Code itself is structured as a distribution of agency" (Mackenzie, 2006, p. 19).

What are traces?

“Researchers can often access traces of social practices in large document repositories, opening a window to patterns of coordination and knowledge work that goes well beyond immediate observations” (Østerlund 2014 et al.)

Traces are the digital artefacts that indicate the existence or passing of something or someone. This can refer to:

  • - system log-files (connection times, IP addresses)
  • - uploaded documents
  • - blog posts
  • - webservice interactions
  • - videos
  • - audio files
  • - code (front or back-end)
  • - chat messages
  • - cronjobs
  • - scores,
  •   or
  • - other information recorded and saved on webservers and databases

  • Assessment story 1

    PIAAC

    "PISA for everybody else"

    The Programme for the International Assessment of Adult Competencies (PIAAC) is run by the OECD and attempts to render quantifiable and commensurable the range of literacy practices of over ‘166,000 people representing more than 724,000,000 adults in 24 countries/economies between the ages of 16-65.

    A network graph depicting the PIAAC network

    Coded scripts

    I am going to ask some questions about ^CurrentLast

     If (B_D12h=1 and B_Q12a=1) then ^CurrentLast=’your participation in open or distance

    education.’ else if (B_D12h=1 and B_Q12c=1) then ^CurrentLast=’your participation in organised sessions for on-the-job training or training by supervisors or co-workers.’ else if (B_D12h=1 and B_Q12e=1) then ^CurrentLast=’your participation in seminars or workshops.’ else if (B_D12h=1 and B_Q12g=1) then ^CurrentLast=’your participation in courses or private lessons.’ else if (B_D12h=2 or B_D12h=4) then ^CurrentLast=’the activities you just reported on. I will start with some questions on the last of these activities that you participated in.’

    INTERVIEWER: Press <Next key> to continue 

    ROUTING: If (B_D12h = 1 and B_Q12c = 1), go to B_Q14b else if (B_D12h = 1 and (B_Q12c = 2 or

    B_Q12c = DK or B_Q12c = RF)), go to B_Q14a else go to B_Q13

    OECD, 2010, p. 27)

    Space

    Translating skills into capital

    ln yin n + β1Cin+ β2 (CinNn ) + β3Ein + β4 E2in+ β5 Gin + εin     

    ln yin = the hourly wage of the individual (i)

     μn = country fixed effects (n=1,..., 22 countries)

    β1 = earnings gradient associated with measured human capital

    Cin = individual skill measure

    β2 = coefficient of interest

    Nn= various measures of country characteristics

    β3Ein = years of actual labour-market experience

    β4 E2in

    β5 Gin= gender indicator

    εin   = stochastic error

     

    Assessment story 2

    Organisation Z

    Organisation Z (a pseudonym) is a network of language schools in France, Spain, and Germany. The head office, situated in Paris, has for its primary activity the provision of Vocational Education and Training (VET) to adults (18-65 years). It positions itself as a node in the distribution and organisation of learning software and teachers of English to Speakers of Other Languages (ESOL).


    Map of depicting interactions between teachers, a learner (Sue), send-email.php, and Shareable Content Objects (2012).

    Same network with more actants

    Map of depicting interactions between teachers, a learners, and Shareable Content Objects (2012).

    SCORM communication

    Tracking Learners

    function sendScore(score, min, max) {

    •  // NB : toString is necessary to keep float values from being rounded.

    •  doLMSSetValue(“cmi.core.score.raw”, score.toString());

    •  doLMSSetValue(“cmi.core.score.min”, min.toString());

    •  doLMSSetValue(“cmi.core.score.max”, max.toString());

    •  doLMSSetValue(“cmi.core.lesson_status”,”completed”); 

    Discussion

    "A focus on the actors, collectives and things in the making and not only on the numbers and ratios" (Boulanger et al., 2014, p. 23)

    ***

    The logic of sampling Pixelated assessment (van der Vleuten, 2014) and the PIAAC sample design argument (IEA-ETS Research Institute, 2014)

    Assessment stories 1 & 2


    *

    €80 000 000 of Personal Training Accounts

    Education & Skills Online has been developed as an assessment relevant to youth and adults of all ages. Institutions, organisations or local governments can use the online tool to assess the skills of a particular population with the goal of providing training or for research purposes.


    References

    Brink, Roelien, & Lautenbach, Geoffrey. (2011). Electronic assessment in higher education. Educational Studies, 37(5), 503-512. doi: 10.1080/03055698.2010.539733

    Central Statistics Office Ireland. (2012). Blaise and PIAAC: Building a survey management system for PIAAC with Blaise. Paper presented at the IBUC 2012 14th International Blaise Users Conference, Strand Palace Hotel, London. http://www.blaiseusers.org/2012/ppt/03b.pdf

     Desrosières, Alain. (2006). From Cournot to Public Policy Evaluation: Paradoxes and Controversies involving Quantification. Prisme, 7. 

    Geiger, Stuart, & Ribes, David. (2011). Trace ethnography: Following coordination through documentary practices. Paper presented at the 44th Hawaii International Conference on System Sciences (HICSS), Hawaii.

    Hanushek, Eric, Schwerdt, Guido, Wiederhold, Simon, & Woessmann, Ludger. (2013). Return to Skills around the World: Evidence from PIAAC.  Paris: OECD.

    Harman, Graham. (2009). Prince of Networks: Bruno Latour and Metaphysics. Melbourne: re.press.

    Latour, Bruno. (2014). An Inquiry into Modes of Existence. An Anthropology of the Moderns. Glossary. from http://www.modesofexistence.org/inquiry/#a=START+UP&s=0

    Østerlund, Carsten, Sawyer, Steve, Ribes, David, Shankar, Kalpana, & Geiger, Stuart. (2014, 4-7 March). What to Do with all those Traces People Leave behind: Computing, Culture, and (Bits of) Context? Paper presented at the iConference 2014, Berlin.

    Mackenzie, Adrian. (2006). Cutting Code. Software and Sociality. New York: Peter Lang.

    Maddox, Bryan. (2014). Globalising assessment: an ethnography of literacy assessment, camels and fast food in the Mongolian Gobi. Comparative Education, 1-16. doi: 10.1080/03050068.2013.871440

    Mol, Anne Marie. (1999). Ontological politics. In J. Law & J. Hussard (Eds.), Actor Network Theory and After (1st ed.). Oxford: Blackwell Publishers.

    OECD. (2010). PIAAC Background questionnaire MS version 2.1 d.d. 15-12-2010.  Paris: OECD.

    OECD. (2013). Skills Outlook Key Findings. Paris: OECD.

    Ribes, David, Jackson, Steven, Geiger, Stuart, Burton, Matthew, & Finholt, Thomas. (2013). Artifacts that organize: Delegation in the distributed organization. Information and Organization, 23(1), 1-14. doi: 10.1016/j.infoandorg.2012.08.001

    Simondon, Gilbert. (1980). On the Mode of Existence of Technical Objects (N. Mellamphy, Trans.): University of Western Ontario.

    Stödberg, Ulf. (2011). A research review of e‐assessment. Assessment & Evaluation in Higher Education, 1-14. doi: 10.1080/02602938.2011.557496

    Venturini, T., Baya Laffite, N., Cointet, J. P., Gray, I., Zabban, V., & De Pryck, K. (2014). Three maps and three misunderstandings: A digital mapping of climate diplomacy. Big Data & Society, 1(2). doi: 10.1177/2053951714543804

     Williamson, Ben. (2014). New Centers of Data Visualization in Education.   Retrieved June 26, from http://dmlcentral.net/blog/ben-williamson/new-centers-data-visualization-education

    Zabal, Anouk , Martin, Silke, Massing, Natascha, Ackermann, Daniela, Helmschrott, Susanne, Barkow, Ingo, & Rammstedt, Beatrice. (2014). PIAAC Germany 2012: Technical Report. Münster/New York: The Federal Ministry of Education and Research.

     


    A special thanks to the University of Edinburgh's Centre for Research in Digital Education for organising this talk.