METHODS USED IN THE MISSISSIPPI POLL

2014 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Laboratory (SRL) of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of two large rooms of telephones, though the Mississippi Poll only employed at most eleven telephone stations at a time. The SSRC is directed by Dr. Arthur Cosby, and the SRL is directed by Dr. John Edwards. The Director of the Survey was Dr. Stephen Shaffer in the MSU Political Science and Public Administration Department, who also produced an on-line summary of the results. SRL Director John Edwards was of invaluable assistance in the sophisticated methodology of incorporating cell phones in the sampling frame, as well as of accurately providing the weighting framework, and of providing the impressive facilities of the SRL for the student interviewers. As with all of the Mississippi Polls, the MSU IRB approved the project as protecting human subjects, those selected for interviewing.

A random sampling technique was used to select the household phone numbers and the cell phone numbers, and the adult 18 years of age and older who first answered the phone was interviewed. Up to ten callbacks were made, and calls were made from 5:00 p.m. to 9:00 p.m. on weeknights (except Fridays), and from 10:00 a.m. to 2:00 p.m. on Saturdays and 1PM to 5PM on Sundays. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data. Twenty undergraduate students in Dr. Shaffer’s Political Analysis class made the phone calls. Three hundred and fifty adult Mississippi residents were interviewed from April 7-30, 2014. The sampling frame included both cell phones as well as household phones, and interviewers used manual dialing. The response rate was 31% (completions as a percentage of both completions and refusals). With 350 people surveyed, the sample error for the full sample was plus or minus 5.4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 5.4%.

Some analyses examine only "likely voters" in the November general election, derived from a likely voter additive scale of the campaign interest, intent to vote in November, and knowledge of the name of the U.S. Representative question items. Sample error for the likely voters in the general election sample was plus or minus 5.8%.

In order to achieve a representative sample, the dataset was adjusted or weighted by selected demographic characteristics to adequately represent the diversity of the entire population, as well as by selection probabilities reflecting those adults having both cell phones and landlines and of multi-adult households.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

2012 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Laboratory (SRL) of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of two large rooms of telephones, though the Mississippi Poll only employed at most twelve telephone stations at a time. The SSRC is directed by Dr. Arthur Cosby, and the SRL is directed by Dr. John Edwards. The Director of the Survey was Dr. Stephen Shaffer in the MSU Political Science Department, who also produced this on-line summary of the results. SRL Director John Edwards was of invaluable assistance in the sophisticated methodology of incorporating cell phones in the sampling frame, as well as of accurately providing the weighting framework, and of providing the impressive facilities of the SRL for the student interviewers. As with all of the Mississippi Polls, the MSU IRB approved the project as protecting human subjects, those selected for interviewing.

A random sampling technique was used to select the household phone numbers and the cell phone numbers, and the adult 18 years of age and older who first answered the phone was interviewed. Up to ten callbacks were made, and calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 6:00 p.m. on Saturdays and 1PM to 9PM on Sundays. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data. Eighteen undergraduate students in Dr. Shaffer’s Political Analysis class made the phone calls. Four hundred and thirty nine adult Mississippi residents were interviewed from April 2-25, 2012. The sampling frame included both cell phones as well as household phones, and interviewers used manual dialing. The response rate was 26%. With 439 people surveyed, the sample error for the full sample was plus or minus 4.8%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4.8%.

Some analyses examine only "likely voters," who consist of 323 individuals in the sample, which results in a sample error level of 5.6% for results in those analyses. Likely voters are determined by responses to an additive scale of the campaign interest, intent to vote in November, and knowledge of one’s U.S. Representative question items.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

2010 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of twenty telephones. The SSRC is directed by Dr. Arthur Cosby, and the Survey Research Unit (SRU) is directed by Dr. John Edwards. He is assisted by the founder of the SRU, Dr. Wolfgang Frese. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and up to ten callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights (Mondays, Wednesdays, and Fridays this year), and from 10:00 a.m. to 6:00 p.m. on Saturdays and 1PM to 9PM on Sundays. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data. Thirty undergraduate students in the Political Analysis class made the phone calls.

Six hundred and four adult Mississippi residents were interviewed from April 5-25. The response rate was 42%. The results were adjusted by demographic characteristics. With 604 people surveyed, the sample error is plus or minus 4.1%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4.1%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduates. SRU Director John Edwards, assisted by sociology Wolf Frese, was of great assistance in creating the CATI questionnaire and providing the required survey expertise.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

2008 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of twenty telephones. The SSRC is directed by Dr. Arthur Cosby, and the Survey Research Unit (SRU) is directed by Dr. John Edwards. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and up to ten callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 1:00 p.m. to 5:00 p.m. on Saturdays and 1PM to 9PM on Sundays. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data. Twenty-two undergraduate students in the Political Analysis class made the phone calls.

Five hundred twenty eight adult Mississippi residents were interviewed from April 2-22. The response rate was 40%. The results were adjusted by demographic characteristics. With 528 people surveyed, the sample error is plus or minus 4.4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4.4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

2006 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of twenty telephones. The SSRC is directed by Dr. Arthur Cosby, and the Survey Research Unit was directed by Dr. Wolfgang Frese. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and six callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 1:00 p.m. to 5:00 p.m. on Saturdays and 1PM to 9PM on Sundays. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data. Twenty-four undergraduate students in the Political Analysis class made the phone calls.

Five hundred seventy four adult Mississippi residents were interviewed from April 3-23. The response rate was 50%. The results were adjusted by demographic characteristics. With 574 people surveyed, the sample error is plus or minus 4.2%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4.2%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate. Sociologist Wolfgang Frese was of great assistance in creating the CATI questionnaire and providing the required survey expertise.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

2004 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of twenty telephones. The SSRC is directed by Dr. Arthur Cosby, and the Survey Research Unit is directed by Dr. Wolfgang Frese. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and six callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 1:00 p.m. to 8:00 p.m. on weekends (9 pm on Sundays). A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data. Twenty-one undergraduate students in the Political Analysis class made the phone calls.

Five hundred twenty-three adult Mississippi residents were interviewed from April 5-21. The response rate was 48%. The results were adjusted by demographic characteristics. With 523 people surveyed, the sample error is plus or minus 4.4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4.4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

2002 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of twenty telephones. The SSRC is directed by Dr. Arthur Cosby, and the Survey Research Unit is directed by Dr. Wolfgang Frese. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and six callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 1:00 p.m. to 8:00 p.m. on weekends (9 pm on Sundays). A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

Six hundred eight adult Mississippi residents were interviewed from April 1 to April 14, 2002. The response rate was 50%. The results were adjusted by demographic characteristics. With 608 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

2000 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby, and the Survey Research Unit is directed by Wolfgang Frese. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and ten callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 1:00 p.m. to 8:00 p.m. on weekends (9 pm on Sundays). A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

Six hundred thirteen adult Mississippi residents were interviewed from April 3 to April 16, 2000. The response rate was 49%. The results were adjusted by demographic characteristics. With 613 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

THE 1999 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby, and the Survey Research Unit is directed by Wolfgang Frese. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and ten callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 11:00 a.m. to 9:00 p.m. on weekends. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

Six hundred fifty-nine adult Mississippi residents were interviewed from April 5 to April 14, 1999. The response rate was 60%. The results were adjusted by demographic characteristics. With 659 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

The Director of the Survey was Douglas Feig in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate and graduate majors. Sociologist Wolfgang Frese was of great assistance in creating the CATI questionnaire and providing the required survey expertise.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

1998 Mississippi Poll

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby, and the Survey Research Unit is directed by Wolfgang Frese. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and ten callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 11:00 a.m. to 9:00 p.m. on weekends. The response rate was 64%. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

Six hundred eight adult Mississippi residents were interviewed from April 14 to April 26, 1998. The results were adjusted by demographic characteristics (education, sex, race, adults, phone numbers) to ensure that all social groups were adequately represented in the survey. Census data for 1996 were used for education, and census data from 1990 for race and sex. With 608 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate and graduate majors. Sociologist Wolfgang Frese was of great assistance in creating the CATI questionnaire and providing the required survey expertise.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

1996 Mississippi Poll

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby. A random sampling technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and five callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 60%. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

601 adult Mississippi residents were interviewed from April 9 to April 20, 1996. The results were adjusted by demographic characteristics to ensure that all social groups were adequately represented in the survey. With 601 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate majors, supervised by graduate student Randolph Burnside. Sociologist Wolfgang Frese was of great assistance in creating the dataset and codebook.

WARNING: This dataset should be weighted by WEIGHT in order to be a representative sample of the entire population.

1994 Mississippi Poll

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby. Two stage, random digit dialing technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and unlimited callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 69%. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

620 adult Mississippi residents were interviewed from April 4 to April 13, 1994. The results were adjusted by demographic characteristics to ensure that all social groups were adequately represented in the survey. With 620 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%. In order to achieve a representative sample, the dataset should be weighted by WEIGHT5. The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate majors, supervised by graduate student Kellei Bishop. Sociologist Wolfgang Frese was of great assistance in creating the dataset and codebook.

WARNING: This dataset should be weighted by WEIGHT5 in order to be a representative sample of the entire population.

1992 Mississippi Poll

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby. Two stage, random digit dialing technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and unlimited callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 64%. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

558 adult Mississippi residents were interviewed from April 1 to April 14, 1992. The results were adjusted by demographic characteristics to ensure that all social groups were adequately represented in the survey. With 558 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate majors, supervised by graduate student Lonnie C. Johnson. Faculty and graduate students who assisted with creating the dataset and codebook were Wolfgang Frese, and Bob Lord.

WARNING: This dataset should be wieghted by WEIGHT6 in order to be a representative sample of the entire population.

1990 Mississippi Poll

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby. Two stage, random digit dialing technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and unlimited callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 73%. A Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

Six hundred and one adult Mississippi residents were interviewed from March 26 to April 3, 1990. The results were adjusted by demographic characteristics to ensure that all social groups were adequately represented in the survey. With 601 people surveyed, the sample error is plus or minus 4%, which means that if every Mississippi resident had been interviewed, the results could differ from those reported here by as much as 4%.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate majors, supervised by Steve Grafton. Faculty and graduate students who assisted with creating the dataset and codebook were Wolfgang Frese, and Lynn Whittington.

WARNING: The dataset should be weighted by WGT7 in order to be a representative sample of the entire population.

SEPTEMBER 1988 MISSISSIPPI POLL

Description of Survey

This telephone survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. The SSRC is directed by Arthur Cosby. Two stage, random digit dialing technique was used to select the households, and a random method was employed in sampling each individual within the household. No substitutions were permitted, and unlimited callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 73%. A Computer Assisted Telephone Interviewing System (CATI) was used to gather the data.

One thousand twenty-two adult residents across the state of Mississippi were interviewed from August 29 to September 17, 1988. The sample error with 1022 adult residents is 3.8%. Data should be weighted by number of adults in the household, number of telephone numbers, education, and race to achieve a representative sample as determined by census data.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate majors. Faculty and graduate students who assisted with creating the dataset and codebook were Frank Howell, Wolfgang Prese, Betsy Waits, and Lynn Whittington.

WARNING: This dataset should be weighted by WEIGHT5 in order to be a representative sample of the entire population.

THE APRIL 1988 MISSISSIPPI POLL - STATEWIDE SAMPLE

Description of Survey

This 1988 telephone poll was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. Two stage, random digit dialing was employed to obtain the sample of households, and random sampling was also used to obtain one interviewee in each household. No substitutions were permitted, and unlimited callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 73%. The Computer Assisted Telephone Interviewing System (CATI) was also used.

Six hundred thirty-two adult Mississippians from across the state were surveyed from April 11 to 24. The sample should be weighted (adjusted) by education, race, and sex so that the size of all groups in the sample match that of the population, correcting for the fact that some groups are less likely than others to own telephones. The poll's sample error was 5%, which means that if all adult residents in the state had been surveyed the results would not have differed significantly by more than 5%.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was sponsored by the SSRC, an organization directed by Arthur Cosby. The survey was conducted by MSU political science undergraduate majors. Faculty and graduate students who assisted with creating the dataset and codebook were Frank Howell, Wolfgang Prese, Betsy Waits, and Lynn Whittington.

WARNING: This dataset should be weighted by WGT6 in order to be a representative sample of the entire population.

1986 Mississippi Poll

Description of Survey

This telephone poll interviewed 611 adult Mississippi residents from throughout the state from February 17 to 26. The survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit simultaneous use of ten telephones. Two stage, random digit dialing was employed to obtain the sample of households, and random sampling was also used to obtain one interviewee in each household. No substitutions were permitted, and unlimited callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 73%.

Since some Mississippians do not own telephones, the sample should be "weighted" by demographic characteristics so that all demographic groups are represented in the sample in proportion to their presence in the population. The sample error with a sample size of 611 respondents is 5% at the 95% confidence level.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. Sponsor of the survey was SSRC, directed by Arthur Cosby. The survey was conducted by MSU political science undergraduate majors. Faculty and graduate students who assisted with creating the dataset and codebook were Frank Howell, Wolfgang Frese, Betsy Waits, and Lynn Whittington.

WARNING: This dataset should be weighted by WGT1 in order to be a representative sample of the entire population.

1984 Mississippi Poll

Description of Survey

This statewide telephone survey was conducted among 610 adult Mississippi residents from April 15 to 29, 1984, which results in a 5% sample error at the 95% combined level. The survey was conducted by the Survey Research Unit of the Social Science Research Center at Mississippi State University with facilities that permit the simultaneous use of ten telephones. Two stage, random digit dialing was employed to obtain the sample of households, and random sampling was also used to obtain one interviewee in each household. No substitutions were permitted, and unlimited callbacks were made. Calls were made from 5:00 p.m. to 9:00 p.m. on weeknights, and from 10:00 a.m. to 9:00 p.m. on weekends. The response rate was 73%.

All analyses should use a weighted sample adjusting for household size, number of telephone numbers in the household and demographic characteristics, so that the sizes of various demographic groups in the sample will approximate their presence in the population. The unweighted sample underrepresented blacks by 7%, the under $10,000 income group by 7%, and high school dropouts by about 9%, largely because these groups are less likely to have telephones. All other demographic groups examined were present in the sample within sample error of their presence in the population.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. The survey was conducted by MSU political science undergraduate majors. Faculty and graduate students who assisted with creating the dataset and codebook were Arthur Cosby, Frank Howell, Wolfgang Frese, Betsy Waits, and Lynn Whittington.

WARNING: This dataset should be weighted by WEIGHT4 in order to be a representative sample of the entire population.

1982 Mississippi Poll

Description of Survey

The statewide survey was conducted from September 7-18. It sampled 894 Mississippi residents 18 years of age and older, so the sample error was + 4% at the 95% confidence level. A subset of 613 likely voters was asked the Senate election items, and sample error for that subset was + 5%. The samples should be weighted by household size, the inverse of the number of different telephone numbers serving the household, and various demographic group characteristics to ensure that each person in the population has an equal probability of being selected in the sample. In the unweighted samples, blacks were underrepresented by 6%, high school dropouts were underrepresented by 12%, the under 25 age group was underrepresented by 3%, and the over 69 group was underrepresented by 3%. All other age, sex, income, and geographic groups are represented in proportion to their existence in the population. Using weighted data adjusts for underrepresented groups, so that the proportions of demographic groups in the weighted sample are virtually identical to their strength in the population.

We employed a telephone survey design and selected numbers through the process of two stage, random digit dialing. One respondent within each household was selected randomly using a modification of the Troldahl-Carter method, and no substitution was permitted. Interviews were conducted from the Survey Research Unit of the Social Science Research Center at Mississippi State University, using facilities that permit simultaneous use of ten telephones. Calls were made from 5 to 9 on weeknights, and from 10 AM to 9 PM on weekends. The response rate was 70%, and unlimited callbacks were made to complete interviews.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. A sponsor of the survey was the Clarion Ledger/Jackson Daily News. The survey was conducted by MSU political science undergraduate majors, supervised by Dr. Douglas Feig. Faculty and graduate students who assisted with creating the dataset and codebook were Frank Howell, Betsy Waits, and Lynn Whittington.

WARNING: This dataset should be weighted by WGT4 in order to be a representative sample of the entire population.

1981 Mississippi Poll

Description of Survey

The survey was conducted from October 19 to November 19, 1981. We employed a telephone survey design and selected numbers through the process of two stage, random digit dialing. One respondent within each household was selected randomly using a modification of the Troldahl-Carter method, and no substitution was permitted. Interviews were conducted from the Survey Research Unit of the Social Science Research Center at Mississippi State University. Calls were made from 5 to 9 on weeknights, and from 10AM to 9PM on weekends. The response rate was 75%, and unlimited callbacks were made to complete interviews. The number of people interviewed was 616, which results in a sample error of + 5% at the 95% confidence level. This means that we can be 95% confident that the percentages that we report are within + 5% of what we would have obtained if we had interviewed all adult Mississippi residents. The date set should be weighted by WEIGHT 2, which adjusts for household size and the inverse of the number of different telephone numbers, as well as the size of such demographic groups as race, gender, and education in the population. The Weight ensures that the weighted sample is representative of the population in terms of key demographic groups. The unweighted sample was relatively representative of the demographic groups in Mississippi, being within 1% of the 1980 census estimates for age, sex, and geographic groups. Blacks were underrepresented by 4%, which is still relatively slight and can be explained by some low income blacks not owning telephones.

The Director of the Survey was Stephen Shaffer in the MSU Political Science Department. A sponsor of the survey was the Office of Graduate Studies and Research. The survey was conducted by MSU political science undergraduate majors. Faculty and graduate students who assisted with creating the dataset and codebook were Frank Howell, Betsy Waits, and Lynn Whittington.

WARNING: This dataset should be weighted by WEIGHT2 in order to be a representative sample of the entire population.