Theme Codes: The OCM

Shelley used a modified version of the Outline of Cultural Materials, or OCM, to code her field notes. The OCM was developed originally by G. P. Murdock in 1950 as a way to index and organize ethnographic materials in the Human Relations Area Files (more about HRAF and the OCM in the section on content analysis in chapter 19). The OCM has gone through several editions over the years, and the latest edition is available online at (Murdock et al. 2004 [1961]).

There are 91 big cultural domains in the OCM, in blocks of 10, from 10 to 91. Block 58, for example, covers marriage with codes for nuptials (585), divorce (586), and so on. Other major domains are things like kinship, entertainment, social stratification, war, health, sex, and religious practices. Every project is unique, so you’ll need codes that aren’t in the OCM, but you can add decimals (or words) and extend the codes forever. Table 13.1 shows Shelley’s adaptation of the OCM code 757 (medical therapy):

Don’t be put off by the lengthiness of the OCM coding list. That is its strength. Lots of anthropologists have used the OCM over the years to code their field notes and other materials. George Foster coded his 50 years of notes on the Mexican community of Tzin- tzuntzan using the OCM, and Robert V. Kemper uses it to code the data that he’s collecting on the same community (George Foster, personal communication; Kemper 2002:289). John Honigman used it in his fieldwork on Canadian Indians; the 37 field researchers in Clyde Kluckhohn’s Comparative Study of Values in Five Cultures Project all used the OCM to code their notes, as did the fieldworkers on the Cornell University team project studying a village in India (Sanjek 1990:108, 232, 331). You’ll only use a fraction of the codes on any given project, but once you start using it in the field, you’ll quickly find yourself building supplemental coding schemes to fit your particular needs.

Figure 13.2 shows how Gordon Gibson (of the Smithsonian) used the OCM to code a series of ethnographic films on the Himba, a cattle-herding society in Namibia (in Kreiss and Stockton 1980:287). The film coded in figure 13.2 is about a wedding, so each piece is coded 585, the OCM code for nuptials. Where the hut is seen, the code for dwellings (342) is inserted. Where the film shows people eating meat, the codes 262 (diet) and 264

FIGURE 13.1.

Field notes from Gene Shelley's study of kidney disease patients (1992). Reproduced by permission.

(eating) are inserted. In the frame at 11:45:10, the code for visiting (574) appears, and in the two earlier frames, the code for childhood activities (857) is inserted. When Gibson did this work in the 1970s, the database was held on a mainframe. Today, you would just watch an ethnographic film on your computer and enter codes as you go.

Many people find the use of number codes distracting. Matthew Miles and Michael Huberman (1994), authors of a wonderful book on qualitative data analysis, advocated

BOX 13.2


I want to make clear the three different uses of the word ''code.'' When I say: ''Use codes for places and informant names,'' the word ''code'' means an encryption device. The object is to hide information, not dispense it. When William Partridge interviewed cannabis growers in Colombia, he identified the texts by a letter code and kept the only copy of his notes in a locked trunk (Kimball and Partridge 1979:174). You don't have to be interviewing cannabis growers to be paranoid about keeping your informants' identity secret. You never know what seemingly innocuous information might embarrass or hurt someone if your data fall into the wrong hands.

When I say: ''Code your notes for the themes you develop in your analysis,'' the word ''code'' means an indexing device. Suppose you do 100 interviews with women about their birthing experience. If you stick the code PAIN into the text whenever a woman mentions anything about pain or about feeling hurt, you'd be using the code PAIN as an indexing device—that is, as a way to find your way back to all the places in the text where anything about pain is mentioned. It's just like an index to a book. It says that ''sampling'' is on page 237 and sure enough, when you go to page 237, you find you're reading about sampling.

The third meaning of the word ''code'' is a measurement device. Suppose you make judgments about the amount of pain—by counting words like ''agony'' as indicating more pain than words like ''distress'' or by looking at the content and meaning of the text and counting ''It was painful, but I got through it'' as indicating less pain than ''I prayed I would die.'' You might use codes like LO-PAIN, MID-PAIN, or HI-PAIN and in this case, you'd be using codes for more than nominal measurement.

the use of words or mnemonics that look like the original concept. Like many researchers, they find that mnemonic codes (like ECO for economics, DIV for divorce, and so on) are easier to remember than numbers. Figure 13.3 shows an example of how to do this.

Another value of using your own codes is that they develop naturally from your study and you’ll find it easy to remember them as you code your notes each day. Strauss and

Table 13.1 Shelley's (1992) Adaptation of the OCM Code 757 on Medical Therapy






CAPD (peritoneal dialysis)


Home dialysis


Adjustment to dialysis


Compliance with medical regime


Machinery involved in dialysis




Medical test results


HIV test results

FIGURE 13.2.

Gibson's coding of the Himba films.

SOURCE: L. Kreiss and E. Stockton, ''Using the Outline of Cultural Materials as a Basis for Indexing the Content of Ethnographic Films,'' Behavior Science Research, Vol. 15, pp. 281-93,1980.

Corbin (1990:68) recommend in vivo codes as names for things. In vivo codes are catchy phrases or words used by informants. In his study of Alaskan fishermen, Jeffrey Johnson heard people talking about a “clown.” The word turned out to be a terrific label for a type of person found in many organizations. The term emerged in vivo from the mouths of Johnson’s informants. (More on in vivo coding in chapter 19).

If you use your own coding scheme, or if you modify an existing scheme (like the OCM), be sure to write up a verbose codebook in case you forget what ‘‘A5’’ or “EMP” or whatever-cute-abbreviations-you-dreamed-up-at-the-time-you-did-th e-coding mean.

And don’t get too picky when you make up your own codes. Coding is supposed to be data reduction, not data proliferation. Mathew Miles was involved in a big ethnographic project to evaluate six schools. All the researchers developed their own codes and the code list quickly grew to 202 categories of actors, processes, organizational forms, and

FIGURE 13.3.

Coding field notes with mnemonics.

efforts. Each of the six researchers insisted that his or her field site was unique and that the highly specialized codes were all necessary. It became impossible for anyone to use the unwieldy system, and they just stopped coding altogether (Miles 1983:123).

The important thing is not which coding scheme you use, it’s that you code your notes and do it consistently. In most projects, the coding scheme takes shape as the notes are written. The scheme is revised a lot before it becomes stable. Some anthropologists, even those who use the OCM, wait a month or more, to see how their field notes are shaping up, before they think about how to code the notes.

< Prev   CONTENTS   Source   Next >