27.10.2014 Views

The Quick Count and Election Observation

The Quick Count and Election Observation

The Quick Count and Election Observation

SHOW MORE
SHOW LESS

Create successful ePaper yourself

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

CHAPTER SIX: THE QUALITATIVE COMPONENT OF THE QUICK COUNT<br />

86 • <strong>The</strong> lead trainer—Observers must be “trained to the forms.” That is, trainers<br />

have to explain to observers the details about exactly how the forms<br />

are supposed to be used. This team member has to be able to think about<br />

the structure <strong>and</strong> content of the form from the point of view of the observer<br />

<strong>and</strong> to anticipate how the structure <strong>and</strong> content of the forms shape<br />

the training of observers.<br />

Each <strong>and</strong> every proposed<br />

question should<br />

be able to pass a series<br />

of “tests.”<br />

• A data analyst—Someone responsible for analyzing data on election day<br />

must be on the team to consider methodological issues of question construction,<br />

the practical challenges of data transmission <strong>and</strong> data entry,<br />

as well as the interpretive challenges of how the data will be configured<br />

<strong>and</strong> used on election day.<br />

With the team in place, the next task is to work together to make the detailed<br />

decisions about precisely how each question will be formulated. Cumulative experience<br />

with qualitative form construction <strong>and</strong> measurement suggests some useful<br />

rules to follow. In effect, each <strong>and</strong> every proposed question should be able to<br />

pass a series of “tests.” <strong>The</strong>se can be summarized as follows:<br />

• <strong>The</strong> usefulness test—For each proposed question, the analyst should be<br />

able to specify first, why it is critical to have that particular piece of information<br />

quickly, <strong>and</strong> second, precisely how the data from that question<br />

will be used in the analysis. If there is no compelling reason for having<br />

the information quickly, or if it is not clear exactly how the data from the<br />

question will be used, then the question should not be asked.<br />

Validity <strong>and</strong> reliability<br />

are the most serious<br />

sources of non-sampling<br />

error plaguing systematic<br />

observation data.<br />

• <strong>The</strong> validity test—Recall that validity refers to how well an indicator, the<br />

data produced by answers to questions on the form, actually measures<br />

the underlying concept to be measured. Here, the question that needs<br />

a clear answer is: Exactly what concept is being measured by the question?<br />

And, is there a better, more direct, or clearer way to formulate the<br />

question to measure that concept?<br />

• <strong>The</strong> reliability test—Reliability has to do with the consistency of the measurement.<br />

<strong>The</strong> goal is to reduce the variation in responses between<br />

observers, that is, to have independent observers watching the same<br />

event record that event in exactly the same way. When questions are<br />

worded ambiguously observers are more likely to end up recording different<br />

results when independently measuring the same event. Note that<br />

validity <strong>and</strong> reliability are the most serious sources of non-sampling error<br />

plaguing systematic observation data.<br />

• <strong>The</strong> response categories test—Response categories for questions have to<br />

satisfy two minimal conditions. First, the response categories should be<br />

exhaustive. This means that the structure of the response categories<br />

should collectively cover all of the possible meaningful ranges of responses.<br />

Second, response categories have to be mutually exclusive. That is,

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

Saved successfully!

Ooh no, something went wrong!