This report cites findings from many of the separate surveys that have been conducted by the Pew Internet & American Life Project since March 2000. The comparisons in use of the internet by men and women in 2002 and 2005 are derived from combining several survey data sets in each of those years. The total number of respondents included in the 2002 findings was 14,416 and for 2005 was 6,403. The 2005 material includes surveys through June of this year. In each of those larger data sets the margin of error is less than plus or minus 2%.
As a general rule the findings issued by the Project are based on the findings of a daily tracking survey on Americans’ use of the Internet conducted by Princeton Survey Research Associates. Our standard poll is in the field for a month and aims to complete about 2,200 interviews. For results based on the total sample of any given monthly sample of that size, one can say with 95% confidence that the error attributable to sampling and other random effects is plus or minus 2.2 percentage points.
The number of internet users has grown in our samples over the years as more and more Americans embrace the internet. Usually the margin of sampling error is plus or minus 3 percentage points.
The sample for each of our tracking surveys is a random digit sample of telephone numbers selected from telephone exchanges in the continental United States. The random digit aspect of the sample is used to avoid “listing” bias and provides representation of both listed and unlisted numbers (including not-yet-listed numbers). The design of the each sample achieves this representation by random generation of the last two digits of telephone numbers selected on the basis of their area code, telephone exchange, and bank number.
In tracking surveys, new sample is released daily and was kept in the field for at least five days. The sample was released in replicates, which are representative subsamples of the larger population. This ensures that complete call procedures are followed for the entire sample. At least 10 attempts are made to complete an interview at sampled households. The calls are staggered over times of day and days of the week to maximize the chances of making contact with a potential respondent. Each household receives at least one daytime call in an attempt to find someone at home. In each contacted household, interviewers ask to speak with the youngest male currently at home. If no male is available, interviewers ask to speak with the oldest female at home. This systematic respondent selection technique has been shown to produce samples that closely mirror the population in terms of age and sex. All interviews completed on any given day are considered to be the final sample for that day.
Non-response in telephone interviews produces some known biases in survey-derived estimates because participation tends to vary for different subgroups of the population, and these subgroups are likely to vary also on questions of substantive interest. In order to compensate for these known biases, the sample data are weighted by form in analysis. The demographic weighting parameters are derived from a special analysis of the most recently available Census Bureau’s Annual Social and Economic Supplement, usually issued in March of each year. This analysis produces population parameters for the demographic characteristics of adults age 18 or older, living in households that contain a telephone. These parameters are then compared with the sample characteristics to construct sample weights. The weights are derived using an iterative technique that simultaneously balances the distribution of all weighting parameters.
Those interested in the sample size of each specific survey can download the crosstab files of those surveys, which are available at: https://www.pewresearch.org/internet/Data-Tools/Download-Data/Data-Sets.aspx, or they can contact Project staff through email@example.com/internet.