ggc540
Feb 22, 2010
Writing Feedback / "Do Phone Calls Increase Voter Turnout?" - article summary and critique [NEW]
I was hoping to get some quick revisions for the rough draft of my article critique, I feel like I'm mostly having trouble with awkward wording and repeating words. I also have no idea what kind of conclusion would be appropriate Any help would be appreciated
Thank you in advance for helping me out.
Article Summary and critique
"Do Phone Calls Increase Voter Turnout?: A Field Experiment." was a 1998 observational study done by Alan S. Gerber, Donald P. Green, two researchers from Yale university's department of political science. Their purpose was to more accurately measure the effects of non-partisan telephone campaigns encouraging people to vote on voter turnout. Before this study, drawing from the fact that there are firms that specialize in the work and and from theories presented by other non-experimental studies, it was widely accepted that phone campaigns are effective in increasing voter turnout.
The authors built upon these earlier theories and hypothesized that it did increase turnout, their goal was to fill in gaps left by small sample sizes and a lack of a control audience of previous studies and find the extent of the phone calls influence, what they found, however, was very different from what they were expecting.
The study was implemented in West Haven CT, A city of 54,000 with an average age of 37.2(above the national average) and an above average representation of minorities.
They began by taking from public records, a list of all registered voters and sorting it by address. They then excluded households with more than two registered voters and P.O. boxes. and submitted the list to a company that provided telephone numbers for around two thirds of those on the list. Households were then randomly assigned to separate groups. They were left with a sample size of 17,100 voters. A professional agency was used to make the calls.
6843 voters were assigned to a control group that received no calls, 8824 voters were assigned to the experimental group which was split up into two parts. One group group received a call that included a statement encouraging them to vote while the other group was asked if they could be counted on to vote. The sizes of these groups were 3595 and 5229 respectively. A third 1433 voter group was also added, In this group they were encouraged to donate blood for the red cross. The authors justify this group by intending to use it as another "benchmark for comparison"
After the election they used voter cross-off sheets to determine who had voted. They then subtracted the turnout rate of the control from the turnout of those in the experimental group and divided dividing difference by the contact rate to isolate the effect of calls.
When they analyzed the data, the authors were in for a surprise, The phone campaign did not have any conclusive effect on voter turnout. There was actually a higher turnout among that received the non-political blood drive call. These turnout rates were 67.1 for those contacted with the political message and 67.8 for those contacted with the blood drive message. The control group experienced a turnout of 53.5, almost the same as the average of all the voters included in the study. It should also be noted that the turnout rate among those included in the study that were not available to be contacted was lower than the average, with turnout rates of 42.0 and 43.5 for the message encouraging them to vote and the blood donation message.
This was a well done study that was effective in providing well founded data that is very applicable to the real world and avoiding the pitfalls of earlier studies on its subject. The precautions the researchers took to improve on those earlier studies are what gives this one its strength.
The sample size of 17,100 provides a much better representation of their population than most earlier studies and even the Adams and Smith study they mention which included 2650. This also helped lower the high standard error that other studies suffered from.
The addition of the third group that received the blood drive call was instrumental in preventing the bias experienced by other researchers when it came to voters that were easier to contact being more likely to vote. They further remedied this problem with the use of some deft statistical procedures that took into account the contact rates unlike other studies, such as the Adams and Smith study which directly compared the turnout of those contacted and not contacted and did not figure/adapt for the possibility of such a bias.
The basics of their experiment were also built on a solid foundation to maintain the external validity of their experiment with the use of random samples and the use of methods and tools that are commonly used in the real world to conduct these campaigns.
While the results of the data are quite solid, the experiment was geographically limited. By only using one city they did not take into account how the location could affect peoples responsiveness to phone calls. Some of the demographics in the city of West Haven could also affect the results, while most are fairly average the average age and minority representation are above average. Older voters could respond differently to phone messages and so could minorities. The fact that the messages unnecessarily mention the League of women voters could affect the turnout of those contacted depending on their disposition towards the organization
I was hoping to get some quick revisions for the rough draft of my article critique, I feel like I'm mostly having trouble with awkward wording and repeating words. I also have no idea what kind of conclusion would be appropriate Any help would be appreciated
Thank you in advance for helping me out.
Article Summary and critique
"Do Phone Calls Increase Voter Turnout?: A Field Experiment." was a 1998 observational study done by Alan S. Gerber, Donald P. Green, two researchers from Yale university's department of political science. Their purpose was to more accurately measure the effects of non-partisan telephone campaigns encouraging people to vote on voter turnout. Before this study, drawing from the fact that there are firms that specialize in the work and and from theories presented by other non-experimental studies, it was widely accepted that phone campaigns are effective in increasing voter turnout.
The authors built upon these earlier theories and hypothesized that it did increase turnout, their goal was to fill in gaps left by small sample sizes and a lack of a control audience of previous studies and find the extent of the phone calls influence, what they found, however, was very different from what they were expecting.
The study was implemented in West Haven CT, A city of 54,000 with an average age of 37.2(above the national average) and an above average representation of minorities.
They began by taking from public records, a list of all registered voters and sorting it by address. They then excluded households with more than two registered voters and P.O. boxes. and submitted the list to a company that provided telephone numbers for around two thirds of those on the list. Households were then randomly assigned to separate groups. They were left with a sample size of 17,100 voters. A professional agency was used to make the calls.
6843 voters were assigned to a control group that received no calls, 8824 voters were assigned to the experimental group which was split up into two parts. One group group received a call that included a statement encouraging them to vote while the other group was asked if they could be counted on to vote. The sizes of these groups were 3595 and 5229 respectively. A third 1433 voter group was also added, In this group they were encouraged to donate blood for the red cross. The authors justify this group by intending to use it as another "benchmark for comparison"
After the election they used voter cross-off sheets to determine who had voted. They then subtracted the turnout rate of the control from the turnout of those in the experimental group and divided dividing difference by the contact rate to isolate the effect of calls.
When they analyzed the data, the authors were in for a surprise, The phone campaign did not have any conclusive effect on voter turnout. There was actually a higher turnout among that received the non-political blood drive call. These turnout rates were 67.1 for those contacted with the political message and 67.8 for those contacted with the blood drive message. The control group experienced a turnout of 53.5, almost the same as the average of all the voters included in the study. It should also be noted that the turnout rate among those included in the study that were not available to be contacted was lower than the average, with turnout rates of 42.0 and 43.5 for the message encouraging them to vote and the blood donation message.
This was a well done study that was effective in providing well founded data that is very applicable to the real world and avoiding the pitfalls of earlier studies on its subject. The precautions the researchers took to improve on those earlier studies are what gives this one its strength.
The sample size of 17,100 provides a much better representation of their population than most earlier studies and even the Adams and Smith study they mention which included 2650. This also helped lower the high standard error that other studies suffered from.
The addition of the third group that received the blood drive call was instrumental in preventing the bias experienced by other researchers when it came to voters that were easier to contact being more likely to vote. They further remedied this problem with the use of some deft statistical procedures that took into account the contact rates unlike other studies, such as the Adams and Smith study which directly compared the turnout of those contacted and not contacted and did not figure/adapt for the possibility of such a bias.
The basics of their experiment were also built on a solid foundation to maintain the external validity of their experiment with the use of random samples and the use of methods and tools that are commonly used in the real world to conduct these campaigns.
While the results of the data are quite solid, the experiment was geographically limited. By only using one city they did not take into account how the location could affect peoples responsiveness to phone calls. Some of the demographics in the city of West Haven could also affect the results, while most are fairly average the average age and minority representation are above average. Older voters could respond differently to phone messages and so could minorities. The fact that the messages unnecessarily mention the League of women voters could affect the turnout of those contacted depending on their disposition towards the organization