27.10.2014 Views

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

310 Chapter 11<br />

times, and each pair appears n – 2 times. For four items, there are n(n –1)/2<br />

6 pairs; each pair appears twice <strong>in</strong> four triads, and each item on the list<br />

appears three times.<br />

It is all this redundancy that reduces the number of triads needed <strong>in</strong> a triads<br />

test. In a complete set of 84 triads for n<strong>in</strong>e items, each pair of items appears<br />

n 2, or seven times. If you have each pair appear just once (called a lambda<br />

1 design), <strong>in</strong>stead of seven times, then, <strong>in</strong>stead of 84 triads, only 12 are<br />

needed. If you have each pair appear twice (a lambda 2 design), then 24 triads<br />

are needed. For analysis, a lambda 2 design is much better than a lambda 1.<br />

Table 11.1 shows the lambda 2 design for 9 items and 10 items.<br />

TABLE 11.1<br />

Balanced Incomplete Block Designs for Triad Tests Involv<strong>in</strong>g 9 and 10 Items<br />

For 9 items, 24 triads are<br />

needed, as follows:<br />

For 10 items, 30 triads are<br />

needed, as follows:<br />

Items<br />

Items<br />

1, 5, 9 1, 2, 3 1, 2, 3 6, 8, 9<br />

2, 3, 8 4, 5, 6 2, 5, 8 7, 10, 3<br />

4, 6, 7 7, 8, 9 3, 7, 4 8, 1, 10<br />

2, 6, 9 1, 4, 7 4, 1, 6 9, 5, 2<br />

1, 3, 4 2, 5, 9 5, 8, 7 10, 6, 7<br />

5, 7, 8 3, 6, 8 6, 4, 9 1, 3, 5<br />

3, 7, 9 1, 6, 9 7, 9, 1 2, 7, 6<br />

2, 4, 5 2, 4, 8 8, 10, 2 3, 8, 9<br />

1, 6, 8 3, 5, 7 9, 3, 10 4, 2, 10<br />

4, 8, 9 1, 5, 8 10, 6, 5 5, 6, 3<br />

3, 5, 6 2, 6, 8 1, 2, 4 6, 1, 8<br />

1, 2, 7 3, 4, 9 2, 3, 6 7, 9, 2<br />

2, 4, 8 8, 4, 7<br />

4, 9, 5 9, 10, 1<br />

5, 7, 1 10, 5, 4<br />

SOURCE: Repr<strong>in</strong>ted from Social Science <strong>Research</strong>, Vol. 5, M. L. Burton and S. B. Nerlove, ‘‘Balanced Design<br />

for Triad Tests,’’ p. 5, 1976, with permission from Elsevier.<br />

For 10 items, a lambda 2 design requires 30 triads; for 13 items, it requires<br />

52 triads; for 15 items, 70 triads; for 19 items, 114 triads; and for 25 items,<br />

200 triads. Unfortunately, there is no easy formula for choos<strong>in</strong>g which triads<br />

<strong>in</strong> a large set to select for a BIB. Fortunately, Burton and Nerlove (1976)<br />

worked out various lambda BIB designs for up to 25 items and Borgatti has<br />

<strong>in</strong>corporated BIB designs <strong>in</strong>to Anthropac (1992a, 1992b). You simply tell<br />

Anthropac the list of items you have, select a design, and tell it the number of<br />

<strong>in</strong>formants you want to <strong>in</strong>terview. Anthropac then pr<strong>in</strong>ts out a randomized

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!