服务承诺
资金托管
原创保证
实力保障
24小时客服
使命必达
Essay学霸提供Essay，Paper，Report，Assignment等学科作业的代写与辅导，同时涵盖Personal Statement，转学申请等留学文书代写。
Essay学霸将让你达成学业目标Essay学霸将让你达成学业目标Essay学霸将让你达成学业目标Essay学霸将让你达成学业目标私人订制你的未来职场 世界名企，高端行业岗位等 在新的起点上实现更高水平的发展
积累工作经验
多元化文化交流
专业实操技能
建立人际资源圈An Analysis on Strong Polls versus Weak Polls
2023-09-03 来源: 类别: Report范文
An Analysis on Strong Polls versus Weak Polls
In the strong survey, according to the CBS News/ New York Times National Poll, conducted by ICPSR in September, 2012 ( for short, in the following text, this poll is called ICPSR) the property in question is, whether people being surveyed “approve and disapprove the way Barack Obama handling his job as President” (ICPSR). In addition subjects in the survey were given three alternatives to respond to the question. They are: I approve; I disapprove; I have no idea (DK/NA). In the ICPSR, the sample size is 1301 subjects, which is large enough to ensure the generalization would not lapse into hastiness. That is because “A sample size of 1000 will give you a confidence interval of +/-3% 19 times out of 20 (95%) ( Cross Validated). What’s more, the population taking part in the survey was sampled randomly through out the nation. The randomness can be reflected by the fact that random-digit dialing (RDD) was used. In the meantime, 100 telephone numbers was stratified by geographic region, are code, and size of place. The phone numbers include both that of landlines and cell phones. In the household, individual being surveyed were selected by a method initiated by Leslie Kish and modified by Charles Backstrom and Ceral Hursh (ICPSR). The people surveyed were 18 or more than 18 years old. They are from different background. This fact can be reflected in the survey that People vary in the following factors: sex, age, race, marital status, education, income, employment status, religious groups, residential area, political party, political philosophy, whether respondents were supporters of the Tea Party movement, whether respondents had children, and whether respondents thought of themselves as born-again Christians. (ICPSR).
Therefore, the methodology used in the survey makes a strong conclusion. Firstly, justas what have been mentioned above, it’s size is much more than the number 1000 which is necessary for an accurate survey. As we know, the larger the size, the more accurate the representativeness of the samples would be. Thus, the possibility of hasty generalization has been wiped out. Secondly, the people been survey were dialed to respond by the researchers, which means that the possibility of self-selecting is also eliminated. In other words, the extent of people’s willingness to take part in the survey would not affect the representativeness of the survey. Thirdly, ICPSR adopted two ways to increase the representativeness of the survey—RDD and Kish method. As we know, the use of the cellphones is increasing and the use of landline phones is decrease, RDD is an efficient way to make sure that different people’s telephone preference would not bias by the survey. In addition, the distribution of telephone within a family is unequal. While someone might have more than two cellphone, the other might have only one or none. However, the Kish’s method guarantees the representativeness within the family. What surprises me is the diversity of people being surveyed. As we know, people’s spiritual or economic background might have more or less impact on their choice, this survey guarantee that people from diverse background are selected in the survey. Thus, basing on the above analysis, the survey is strong and representative enough to support its conclusion. In my point of view, the numerical value of the survey worth 8.5. The reason why I did not grade it at 9 or 9.5 is that the survey was conducted in one day, on September 1st of 2012, which would harm the representativeness of the survey. It is possible that the respondents might to too busy to be patient to answer the question on their workdays. Perhaps people’s choice might change in more than one day. Thus, though this survey is strong, it does not worth high than 8.5.
The weak survey, according to “Five Reasons There Are Bad Polls”, is conducted by Quinnipiac University, showing that in Ohio, Pennsylvania, and New Jersey, Barack Obama won a substantial margin. However, the problem is that the poll’s results are basing on their partisan and is thus biased. For example, The poll has the president winning among independent voters in Pennsylvania by 22 points, 58-36(Five Reasons There Are Bad Polls). However, this result is based on biased samples and hasty generalization. According to this comment, there are five reasons why they would biase to this result. Firstly, the polling is conducted in summer. However, as we all know, people in the summer and weekend have no time to pick up their phones. For instance, they might go outside for a long travel, and even travel aboard. How could we imagine people who are travelling we pick and answer tedious question raised by the pollsters? Even assuming people don’t go for a long travel, they might go to churches rather than stay at home, as a result even pollsters are conducting with the RDD method, those who use landlines phone would not have the chance to answer the questions. As a result, the representativeness would be weakened by the wrong time in which they conduct the poll. In another word, people who stay at home might be those old people. However, who are active about the American politics are young people, which mean that those who should take part in the politics lose their way to influence the American politics. In addition, even people know that the result of the poll might contradict with other polls they conducted or other people conducted before, they don’t want to give it up or conduct it again. The reason is, firstly, just as what has been mentioned above, people who conducted this poll like the result of it, since it seems to be in Obama’s favor. Secondly, According to the essay, the polling is too expensive that they don’t want to give it up and repeat it. “Media organizations aren’t rich the way they used to be, and they don’t have the luxury of ditching a bad poll that is simply the result of statistical noise and other difficulties” (Five Reasons There Are Bad Polls)
To sum up, a strong survey is conducted in a neutral and scientific way—with a proper sampling and methods such as RDD to guarantee representativeness, while a bad poll is biased by partisan attitude and bad methods, at a wrong time, for example.
Citation:
CBS News/New York Times National Poll, September #1, 2012 (ICPSR 34632)
http://www.icpsr.umich.edu/icpsrweb/ICPSR/studies/34632
CrossValidated: http://stats.stackexchange.com/questions/166/how-do-you-decide-the-sample-size-when-polling-a-large-population
Five Reasons There Are Bad Polls:
In the strong survey, according to the CBS News/ New York Times National Poll, conducted by ICPSR in September, 2012 ( for short, in the following text, this poll is called ICPSR) the property in question is, whether people being surveyed “approve and disapprove the way Barack Obama handling his job as President” (ICPSR). In addition subjects in the survey were given three alternatives to respond to the question. They are: I approve; I disapprove; I have no idea (DK/NA). In the ICPSR, the sample size is 1301 subjects, which is large enough to ensure the generalization would not lapse into hastiness. That is because “A sample size of 1000 will give you a confidence interval of +/-3% 19 times out of 20 (95%) ( Cross Validated). What’s more, the population taking part in the survey was sampled randomly through out the nation. The randomness can be reflected by the fact that random-digit dialing (RDD) was used. In the meantime, 100 telephone numbers was stratified by geographic region, are code, and size of place. The phone numbers include both that of landlines and cell phones. In the household, individual being surveyed were selected by a method initiated by Leslie Kish and modified by Charles Backstrom and Ceral Hursh (ICPSR). The people surveyed were 18 or more than 18 years old. They are from different background. This fact can be reflected in the survey that People vary in the following factors: sex, age, race, marital status, education, income, employment status, religious groups, residential area, political party, political philosophy, whether respondents were supporters of the Tea Party movement, whether respondents had children, and whether respondents thought of themselves as born-again Christians. (ICPSR).
Therefore, the methodology used in the survey makes a strong conclusion. Firstly, justas what have been mentioned above, it’s size is much more than the number 1000 which is necessary for an accurate survey. As we know, the larger the size, the more accurate the representativeness of the samples would be. Thus, the possibility of hasty generalization has been wiped out. Secondly, the people been survey were dialed to respond by the researchers, which means that the possibility of self-selecting is also eliminated. In other words, the extent of people’s willingness to take part in the survey would not affect the representativeness of the survey. Thirdly, ICPSR adopted two ways to increase the representativeness of the survey—RDD and Kish method. As we know, the use of the cellphones is increasing and the use of landline phones is decrease, RDD is an efficient way to make sure that different people’s telephone preference would not bias by the survey. In addition, the distribution of telephone within a family is unequal. While someone might have more than two cellphone, the other might have only one or none. However, the Kish’s method guarantees the representativeness within the family. What surprises me is the diversity of people being surveyed. As we know, people’s spiritual or economic background might have more or less impact on their choice, this survey guarantee that people from diverse background are selected in the survey. Thus, basing on the above analysis, the survey is strong and representative enough to support its conclusion. In my point of view, the numerical value of the survey worth 8.5. The reason why I did not grade it at 9 or 9.5 is that the survey was conducted in one day, on September 1st of 2012, which would harm the representativeness of the survey. It is possible that the respondents might to too busy to be patient to answer the question on their workdays. Perhaps people’s choice might change in more than one day. Thus, though this survey is strong, it does not worth high than 8.5.
The weak survey, according to “Five Reasons There Are Bad Polls”, is conducted by Quinnipiac University, showing that in Ohio, Pennsylvania, and New Jersey, Barack Obama won a substantial margin. However, the problem is that the poll’s results are basing on their partisan and is thus biased. For example, The poll has the president winning among independent voters in Pennsylvania by 22 points, 58-36(Five Reasons There Are Bad Polls). However, this result is based on biased samples and hasty generalization. According to this comment, there are five reasons why they would biase to this result. Firstly, the polling is conducted in summer. However, as we all know, people in the summer and weekend have no time to pick up their phones. For instance, they might go outside for a long travel, and even travel aboard. How could we imagine people who are travelling we pick and answer tedious question raised by the pollsters? Even assuming people don’t go for a long travel, they might go to churches rather than stay at home, as a result even pollsters are conducting with the RDD method, those who use landlines phone would not have the chance to answer the questions. As a result, the representativeness would be weakened by the wrong time in which they conduct the poll. In another word, people who stay at home might be those old people. However, who are active about the American politics are young people, which mean that those who should take part in the politics lose their way to influence the American politics. In addition, even people know that the result of the poll might contradict with other polls they conducted or other people conducted before, they don’t want to give it up or conduct it again. The reason is, firstly, just as what has been mentioned above, people who conducted this poll like the result of it, since it seems to be in Obama’s favor. Secondly, According to the essay, the polling is too expensive that they don’t want to give it up and repeat it. “Media organizations aren’t rich the way they used to be, and they don’t have the luxury of ditching a bad poll that is simply the result of statistical noise and other difficulties” (Five Reasons There Are Bad Polls)
To sum up, a strong survey is conducted in a neutral and scientific way—with a proper sampling and methods such as RDD to guarantee representativeness, while a bad poll is biased by partisan attitude and bad methods, at a wrong time, for example.
Citation:
CBS News/New York Times National Poll, September #1, 2012 (ICPSR 34632)
http://www.icpsr.umich.edu/icpsrweb/ICPSR/studies/34632
CrossValidated: http://stats.stackexchange.com/questions/166/how-do-you-decide-the-sample-size-when-polling-a-large-population
Five Reasons There Are Bad Polls:
http://www.commentarymagazine.com/2012/08/01/five-reasons-there-are-bad-polls/
Essay学霸网站原创范文除特殊说明外一切图文著作权归Essay学霸所有；未经Essay学霸官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，Essay学霸保留一切法律追诉权。
