How many of these Nielsen Media Research families are Hispanic-American?

Nielsen Media Research has maintained a separate Hispanic-American Service and associated samples since the early 1990s. All the homes in these samples are Hispanic-American. The reasons for the separate service are quality oriented and represent our commitment to measuring the Hispanic-American Audience in the best possible way. To do this, we have committed large resources to establishing a bi-lingual field force to increase cooperation among Hispanic-American households.

In addition, we've added bi-lingual diaries, bi-lingual interviewers, and commissioned a yearly "enumeration" study which determine the language spoken in the home by Hispanic-American homes. This is important since homes which speak only Spanish tend to view differently from those Hispanic-Americans who speak only English. We want to make sure that we have the proper mix of language in our sample.

Let’s look at the individual Hispanic-American samples.

National Hispanic-American People Meter Sample
There are 800 households in this sample. The number fluctuates daily as households come in and out of the sample every month. The entire sample consists of Hispanic-American TV households.

Metered Market Samples
In five of the largest Hispanic-American markets, Nielsen Media Research uses meters for local television measurement. There are a total of approximately 1,000 metered households in these markets (or about 200 households for each market). Of the 1,000 sample homes, all are Hispanic-American. See table below for characteristics of each market.

Diary Samples
Every year Nielsen Media Research mails out more than one million paper diaries to sample homes in the U.S. The diaries are mailed during the so-called sweep months -- a minimum of four times a year. Households are asked to write down what they are watching and who is watching for one week. These samples change all the time. Homes are not asked to keep a weekly diary more than once, and metered households are not asked to keep diaries at all. The Hispanic diaries from our general market sample from Hispanic-American homes are used in the Hispanic-American samples and are supplemented by additional homes if necessary to ensure that each market has a sufficient sample.

But just because a household initially agrees to keep a diary and return it to us, that doesn't mean they will always do so. As a result, sample characteristics often fluctuate. In many instances, the number of diaries returned does not always match the known composition of the audience.   In those instances, however, where the percentage of returned diaries falls below our Universe Estimate by language, Nielsen Media Research "weights" the returned diaries up to the known percent. This procedure assures an even playing field and that the ratings are proportionately representative of the actual language spoken in the Hispanic-American homes.

How can small samples of African-American and Hispanic-American households represent the entire ethnic viewing audience?

Actually, a representative sample doesn’t have to be very large to represent the population it is drawn from. The sample does, however, need to be selected in a way which gives every member of the population a known statistical chance of being chosen. If every member of the population has an equally good chance of being in the sample, then this makes it a representative sample. Through statistical theory (and many years of practical experience which is consistent with that theory), we know that fairly drawn (or random) samples vary in usually small ways from the population. Over time these small differences tend to average out.

Unless the population is very small, a fixed sample size will yield essentially the same degree of statistical reliability regardless of the size of the population. For example, a sample size of 560 African-American households measuring the current African-American population of 12.1 million TV households will yield essentially the same degree of reliability as a total sample of 560 households measuring the total population of 102 million TV households. For comparison purposes, the current installed African-American sample size of 560 households is similar to the New York metered market installed household sample size of 540, which measures a population of 6.7 million TV households.

What techniques does Nielsen Media Research use to collect viewing information on African-American and Hispanic-American Households?

Nielsen Media Research uses people meters, set tuning meters, paper diaries, and telephone coincidentals. We also employ many bi-lingual workers during the sample recruiting process to improve cooperation among Hispanic-American households.

More...