CPLS DATABASE

The CPLS database is a collection of ethnographic case studies of literacy practice in various marginalized cultural communities. The database is intended to both document various literacy practices and facilitate cross-case analyses of these practices. We believe that principled cross-case analyses, facilitated by this database, will allow us to (a) reach for greater generalizability than that afforded by a single case, and (b) deepen understanding and explanation of literacy and the ways that it manifests itself and mediates the lives of people across different cultural contexts.

Development of the Database

This database has undergone several iterations. It began as a simple Excel spreadsheet, yet through this first attempt at a meta-matrix (Miles & Huberman, 1994; Purcell-Gates, 2007), we realized that we could not do much more than 'count instances' across contexts without bringing to bear deep knowledge of each case within which the instances were actualized and imbued with meaning. We then decided to include in our definition of database the qualitative data that informed each case study as well as the researcher interpretations of that data. As a multi-dimensional database, this can be used by researchers with the 'flat database' (LeCompte & Schensul, 1999, p. 127) of theoretically coded literacy events.

Characteristics of the Database

As of February, 2010, the CPLS dataset includes qualitative and quantitative – interpretive and descriptive – data, corresponding to twenty ethnographic case studies that have been carried out in eight different countries. However, the number of case studies included in the dataset is expected to continuously increase, thus constantly adding new data to an expanding and evolving database. The data that composes our dataset then reflect the many different facets of literacy as it is practiced around the world. All data have been provided by the original authors of the case studies. The dataset can be accessed by any CPLS affiliated researcher at any time, for future cross-case analyses. Affiliated researchers retain sole authorship rights of their individual studies and are guaranteed acknowledgement (reference) in any future analyses by other CPLS researchers who use data from their studies.

The CPLS Literacy Practice Database consists of two types of data: (a) The multi-dimensional case study data; and (b) the theoretically coded literacy events contained in a flat dataset. The case study data is stored within Hermeneutic Units (Atlas.ti format). For each study, researchers load into an Atlas.ti Hermeneutic Unit their field notes, interview transcripts, interview protocols, photos of environmental texts found in the communities under study (e.g., signs, books, advertisements, magazines, etc.), scanned artifacts collected during the conduct of the field work (e.g., newspapers, flyers, memos), official documents, and all other collected data such as the demographic information of participants, video recordings, etc. Below are examples of these types of data:

Coded Photographs of Environmental Texts

Atlas.ti allows for coding of various images. This screen shot below shows the code string for an environmental text (a blackboard with a directive to potential customers) observed by a researcher in Vancouver, Canada. The string of codes provides information such as which case study the text came from, the location/context for this text, the purpose for the text, and so forth.

Click to see bigger image

A second example of a coded image: The code string for this environmental text (a banner with a welcome message for potential students) observed by a researcher in Vancouver, Canada. The string of codes provides information such as which case study the text came from, the location/context for this text, the purpose for the text, and so forth.

Click to see bigger image

Coded Interview Transcripts

This screenshot provides an example of how CPLS researchers transcribe and code interviews. In this example, the researcher has identified and coded various literacy events described by the participant. The code string identifies information such as the social activity domain of the participant within which the event happened, the text that was read/written, the social purpose for the reading/writing of that text, and the online reading function for reading/ writing that text.

Click to see bigger image

A second example of a coded interview transcript, in which the researcher has identified and coded various literacy events described by the participant. The code string identifies information such as the social activity domain of the participant within which the event happened, the text that was read/written, the social purpose for the reading/writing of that text, and the online reading function for reading/writing that text.

Click to see bigger image

We also upload final reports, conference presentations, drafts of articles, etc. - all accounts of researcher interpretation of the data. The data are then coded iteratively according to (a) researcher interest/questions; and (b) common literacy event codes and conventions used across the CPLS cases.

The flat database begins in Atlas.ti where the coding is done and then exported to a meta-matrix in text format. This database contains both (a) codes that serve data management and descriptive purposes, and (b) theoretically-based codes for conceptual purposes. Within the main part of the database, we have used a code string with nine code types for each identified literacy event. Among these nine codes, four are conceptually-based and related to our emerging model of a literacy practice while the others are considered more descriptive. (For information on demographic codes, please see in the Working Papers section.)

Data exported from Atlas.ti into Excel spreadsheets

The CPLS team developed a way to import data coded within Atlas.ti into an Excel spreadsheet, which serves as a "flat" database. Although this screenshot only shows a small fraction of this flat database, it shows that each specific code from the code strings in Atlas.ti is imported into the database, allowing for comparisons to be made across cases.

Click to see bigger image

In addition to coded data from Atlas.ti, the "flat" database also includes demographic data about participants from each case study.

Click to see bigger image

Currently, the database includes data from participants with 16 different nationalities, 20 native languages, 56 occupation types, all levels of schooling, 4,555 individual literacy events, 349 different functions of reading/writing, 342 social purposes of reading/writing; texts written in 23 different languages, and 774 different types of texts read or written. These numbers are constantly rising as new data from completed projects are entered.

References

LeCompte, M.D. & Schensul, J.J. (1999). Designing & Conducting Ethnographic Research. Walnut Creek, CA: AltaMira Press.

Miles, M.B., & Huberman, A.M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage.

Purcell-Gates, V. (Ed.). (2007). Cultural practices of literacy: Case studies of language, literacy, social practice, and power. Mahwah, NJ: Erlbaum.