Mitch Charnley dedicated his life to furthering the education of those around him. In his career, Charnley worked for the Honolulu Star Bulletin, Detroit News, and The American Boy among others. In 1934, he joined the University of Minnesota as an assistant professor in journalism and by the time he retired in 1966, after more than 30 years of teaching, his name was forever remembered as a symbol of excellence and ingenuity in the journalism field. His students included Eric Sevareid and Harry Reasoner, and he wrote a number of textbooks, including Reporting and News by Radio. He received the Distinguished Achievement Award from the Radio-TV News Directors Association in 1963, and in 1968 he was awarded Sigma Delta Chi's Distinguished Teaching in Journalism Award.
On Feb. 16, 1991, Charnley passed away, leaving behind a legacy of greatness that has lived on through the Mitchell V. Charnley Alumni Professorship in Journalism, which provides students in the Hubbard School with a chance to be partnered with faculty in a real-world situation that breaks away from a lecture-based class and instead offers an in-depth, hands-on experience.
Fall semester of 2018 was the first time the fund was used to implement the Charnley Project course in which strategic communication and journalism students worked together to investigate and collect data on political ad spending and then in turn publish an article about their findings in MinnPost. In fall 2019, students tackled a new topic: public opinion in the state of Minnesota.
During the Fall 2019 semester, Chris Ison co-taught the Charnley Project course with Assistant Professor Benjamin Toff. The two classes - Communication, Public Opinion, and Social Media taught by Toff, and In-Depth Reporting taught by Ison - had 66 students working side by side to evaluate public opinion in Minnesota. Toff’s class was responsible for creating and sending out surveys of residents in the state and University of Minnesota undergraduates, while Ison’s students took the survey information and further investigated and reported on the responses. Find their work below. More stories will be added in the coming weeks.
- More information on the Minnesota resident survey
The purpose of the survey was to provide students in the course a real-world example of how survey research is conducted as part of journalism about public opinion. The full questionnaire is available upon request (email@example.com).
The poll was conducted for the Hubbard School of Journalism and Mass Communication by Strategic Research Group, which worked with Marketing Systems Group (MSG) to identity residential blocks to reach targeted populations. Results are based on 707 completed responses collected between October 2-31, 2019, although missing data and lack of similar records to fill in gaps in demographic data led to five of these cases being excluded from the final data as they could not be weighted.
Respondents were selected from households using a randomized address-based sample (ABS) approach, which included a targeted oversample of residential blocks expected to contain higher rates of 18-29-year olds and high-density immigrant communities. Despite this oversampling, these populations were still under-represented in the final response data, which used a weighting technique to match the sample to population demographics of Minnesota residents. The demographics utilized in the weighting of the Minnesota Resident Survey were age, gender, education, race, and geography type (urban, suburban, rural).
We defined urban, suburban, and rural areas using zip codes respondents reported for where they lived. Respondents were coded as urban if they lived within the boundaries of the cities of Minneapolis or Saint Paul (in Hennepin and Ramsey counties, respectively) or Duluth, St. Cloud, or Rochester (using lists of zip codes provided by American Public Media in their 2017 Ground Level Survey of Minnesotans). Respondents were coded as suburban if they resided in the remaining areas of Hennepin and Ramsey counties or the surrounding counties that make up the Twin Cities metro area (Anoka, Carver, Dakota, Scott, and Washington counties). All other respondents were coded as rural.
Initial invitation letters sent by mail to 15,000 selected households with follow-up non-response prompting conducted via telephone, postcards, and email for households with valid addresses available. These contacts resulted in a response rate for the survey effort of 4.7 percent. Survey responses were recorded primarily online (97 percent) with small number (3 percent) completing the survey over the telephone.
The margin of sampling error for the complete set of data is ±3.7 percentage points. This means that in 95 out every 100 samples drawn using the same methodology, estimated proportions based on the entire sample are likely to be no more than 3.7 percentage points away from their true values in the population. Margins of error for subgroups within the sample, such as different age groups or Democrats and Republicans, are larger than the margin of error for the total sample. Sampling error does not take into account additional sources of variation in public opinion surveys, such as nonresponse, question wording or other context effects.
A design effect has also been calculated to account for the loss in statistical efficiency that results from systematic differences in non-response among subgroups. The total sample design effect for this survey is 3.09, which results in an additional ±2.8 percentage points added to the margin of sampling error.
- More information on the Minnesota student survey
The full questionnaire is available upon request (firstname.lastname@example.org).
The Minnesota Student Survey was conducted by the Hubbard School of Journalism and Mass Communication on a random sample of current University of Minnesota, Twin Cities, undergraduates whose contact information was provided by the Office of Institutional Research. Results are based on 330 completed responses collected between October 29-November 8, 2019. The sample included survey weights by gender matched to official enrollment statistics to account for small differences in non-response.
Initial invitation emails were sent to 1,900 selected students with two subsequent follow-up emails sent to students who did not initially respond or complete the full questionnaire, resulting in a response rate of 17.4 percent. Survey responses were recorded online using the Qualtrics survey platform.
The margin of sampling error for the complete set of data is ±5.4 percentage points. This means that in 95 out every 100 samples drawn using the same methodology, estimated proportions based on the entire sample are likely to be no more than 5.4 percentage points away from their true values in the population. Margins of error for subgroups within the sample, such as Democrats and Republicans, are larger than the margin of error for the total sample. Sampling error does not take into account
additional sources of variation in public opinion surveys, such as nonresponse, question wording or other context effects.
Pass the pie and shut your piehole
Though most Minnesotans — about seven of every 10 — talk politics with family and friends at least once a week, over a third of have stopped talking with at least one person due to politics, according to a recent survey conducted by the University of Minnesota’s Hubbard School of Journalism and Mass Communication.
In 2018 Assistant Professor Christopher Terry and Teaching Professor Chris Ison co-taught the class, which was the first time the fund was used to implement the Charnley Project course. The students investigated and collected data on political ad spending and then published an article about their findings in MinnPost. Read their work below.