Main Menu

  • Monmouth University Polling Institute Logo


    The Monmouth University Poll’s standard methodology utilizes a random digit dial (RDD) “probability” sampling design to select survey participants. All telephone exchanges (i.e., area code and first three digits of the phone number) that reach a household are programmed into a computer. The remaining four digits of the telephone number are then randomly generated by computer. This random digit dial process ensures that people with unlisted as well as listed numbers are included in the sample.

    After interviewing is complete, the sample is “weighted” to correct for the fact that some respondents are harder to reach than others. For example, younger adults are generally much harder to reach at home, even after multiple call attempts. According to census figures, approximately 18% of the adult population is between the ages of 18 and 29. However, only 12% of the people actually surveyed in a hypothetical poll may fall in that age group. To make sure that the final results of the survey are representative of the population, each person age 18 to 29 in the sample must be counted one and a half times by the computer (12% x 1.5 = 18%) to make sure their prevalence in the sample proportionately represents their prevalence in the population. This is balanced by the fact that other age groups are usually over-represented in the sample and are therefore weighted “down.”

    The Monmouth University Poll either tracks or adjusts for geography, gender, race, age, and education to comport with current U.S. census figures of the adult population. This weighting ensures that findings from a sample can be generalized to the full target population.

    All surveys are subject to “sampling error”—that is the expected probable difference between interviewing everyone in a population and interviewing a scientifically selected sample of that population. Sampling error is based on the size of the sample. This is true whether you are conducting a public opinion survey or whether you are counting how many fish are in a lake by sampling different parts of the lake.

    However, in public opinion polling, sampling error cannot take into account other sources of variation inherent in this type of research, such as non-response, question wording, or context effects. So that interested parties may make their own evaluations, the full text of poll questions are made available in each Monmouth University Poll release.

    (Note: Exceptions to this methodology may occasionally occur if the institute surveys a relatively small or specialized population. However, any departure from our standard sampling methodology will be noted in the relevant polling release).