Discussions
jenni0597
student_response_1.docxStudent response 1
The sampling could lead to bias or error for several reasons, the 3 main ones are: a) the data is not up-to-date: this census was taken in 2010 or before and we are in 2014, that means that these data are old and in some cases, outdated, and therefore may not be reliable depending on what the information is requires for. b) Not all of the persons within a specified area will provide the required information to satisfy the census criteria to be a true representative sample: Americans in general are very sensitive about giving out personal information to anyone, least the government and therefore, it can be expected that some of the information the census personnel received will contain errors. And c) The samplers / data gatherers will not get to meet everyone within their respective sample area to compile any data on then. They may visit a house and find that there is no one there to answer their questions (and while some of them get around this by asking the neighbors basic questions about their next door neighbors). They either leave without any information, limited or inaccurate information (what they get from the neighbors). Are they neighbors that are bias towards their neighbors? What about poll workers? The fact is, errors and biases can hardly be avoided in some cases.
Student response 2
The Census Bureau’s sampling process could lead to a bias or error in the data regarding the population of zip code 30331. The information provided by American FactFinder can be classified as secondary data. “For many years the census has been the backbone secondary data in the United States” (Burns & Bush, 2012). Secondary data provides a number of uses for marketing research and marketing researchers decide what to include or dispel from their studies. “Data needed for marketing management decisions can be grouped into two types: primary and secondary” (Burns & Bush, 2012). Data collected though secondary sources tend to be accessible, inexpensive and enhance primary research that has previously conducted (Burns & Bush, 2012).
Along with the advantages secondary data presents, market researchers must also be weary of the existing disadvantages. Problems associated with secondary information include outdated data, incompatible reporting units and unusable class definition. “These problems exist because secondary data has not been collected specifically to address the problem at hand…” (Burns & Bush, 2012). The American FactFinder shows the 2010 population for various zip codes and housing characteristics. Zip code data can be inadequate for researchers when attempting to ascertain specific information about the market. Furthermore, the census does not account for every individual in an area due to lack of data and incorrect data. Inaccurate data can be detrimental to marketing research efforts as conflicting numbers can result in the misrepresentation of facts.
Burns, A. C. & Bush, R. F. (2012). Basic marketing research using Microsoft Excel data analysis (3rd ed.). Upper Saddle River, NJ: Pearson Prentice Hall.
Student response 3
|
Discussion 2: Samples & Sampling I used the American Fact Finder to look up the most current population of my city which is Upper Marlboro Maryland. They didn't have anything current but the 2010 Census. The population form 2010 was 43,013. The male population is 19,179 and the women is 23,172. The reason why this type of sampling process could lead to an error or it being bias, is because the information is not current data. The numbers could be much higher with gender and age group. You also have to consider all the possible people that didn't complete or participate in the census surveys. You cannot fully go off the data findings of an incomplete survey. Reference http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml
|
Student response 1
When conducting research objectives and Data Analysis, it is important to gather your sampling well in order to come up with a good hypothesis. If you fail to do this, your project may not be credible and the criteria you use could also be questionable. Description:Description research objectives, the researcher will perform summarization analysis, defined as describing the data in the sample with the use of percentages or averages. This is an important element in the process, along with Generalization: That is, if the American Express executives believe that 40% of college students currently own an American Express credit card, the researcher could test his or her sample percent of respondents against 40% to see if the executives' belief is supported or refuted. This generalization turns out to be inaccurate, according to another research; however, it still represents an important step.
Differences: The differences analysis, the researcher identifies a categorical variable (such as gender) and compares the groups represented by that variable (males versus females) by analyzing their differences on a second variable. This can tell you how your research compared with the sample available and last but not least is the Relationships: If the researcher isolates two variables and both are categorical, he or she will perform cross-tabulation analysis, but if the two variables are metric, correlations will be used. (Bush, p.244 & 245) Getting these four elements right can go a long way in providing you as a researcher with the relevant information to make informed decisions.
Bush, Alvin C Burns and Ronald F. Basic Marketing Research with Excel, 3/e for Ashford University, 3rd Edition. Pearson Learning Solutions. VitalBook file.
Student response 2
Description: Percentage of total fatal occupational injures, men and women workers, by event, 2009 (Pg 6)
When data analysis is used to fulfill the objective of description, the sample data is often summarized with percentages and categorical data (Burns & Bush, 2012). Fatal injures occur in the workplace for various reasons which is why employee safety is general held as a top priority. The Bureau of Labor Statistics site uses a bar graph to illustrate the percentages of fatal occupational injuries between male and female workers. Furthermore, The bar graph shows the number of fatal occupational injures and breaks the numbers down by type of fatal events ranging from falls to fires and explosions.
Generalization: Average annual expenditures by single women, by level of income before taxes, selected expenditures, 2008-09 (Pg 4)
When data analysis is used to fulfill the objective of generalization, sample findings are generalized in terms of population data (Burns & Bush, 2012). Generalizing annual expenditures of single women involves representing potential group spending averages. The Bureau of Labor Statistics site views the single women population and generalizes about the amount of money being spent on various expenses. Generalizations are generally inaccurate when it comes to accounting for every dollar spent by the 2008-2009 single women population.
Differences: Ratio of women’s to men’s earnings, selected occupations, 2010 (Pg 7)
When data analysis is used to fulfill the objective of differences, sample data and percents are compared with one another in an effort to find a meaningful discrepancy (Burns & Bush, 2012). Men and women hold similar positions in the workforce, however men and women make different amounts of money in many occupations. The Bureau of Labor Statistics site uses a bar graph to represent the disparity of funds in occupations ranging from postal service clerks to lawyers. The various selected occupations show the ratios of money made between the genders.
Relationships: Average hours per week women spent in selected primary activities, by age and educational attainment, 2009 (Pg 2)
When data analysis is used to fulfill the objective of relationships, sample data and variables are related to each other in an effort to find meaning correlations (Burns & Bush, 2012). Women spend their time doing work related activities, leisure activities, and household activities. The Bureau of Labor Statistics site uses sample data such as age and educational attainment as well as how these elements can impact the amount of time spent on respective activities. By identifying which numbers relate to one another, data analyst gain more insight into the activities of women and how age can impact decisions.
Burns, A. C. & Bush, R. F. (2012). Basic marketing research using Microsoft Excel data analysis (3rd ed.). Upper Saddle River, NJ: Pearson Prentice Hall. Women at Work (2011). U.S. Bureau of Labor Statistics http://www.bls.gov/spotlight/2011/women/pdf/women_bls_spotlight.pdf
Student response 3
There are four different types of research objectives that are explained in our textbook in Table 11.1. These four types are description, Generalization, Differences, and Relationship. We were asked to look at the BLS Spotlight on Statistics: Women at Work and find an example for each research objective in the statistics.
Description is where the sample data is being summarized. An example of this would be Educational Attainment of young Women. “There were 23.4 percent of young women around the age of 23 that held a bachelor’s degree, compared to 14.3 percent of young women. The U.S. Bureau of Labor Statistics uses a bar graph to show the percent for high school dropouts for each age of 18 to 23. It shows how many were enrolled in high school at those ages, high school graduates and GED recipients that were not enrolled in college, enrolled in college, and bachelor’s degree or more for young women.
Generalization is findings in the sample of population; these samples are usually through hypothesis tests or confidence intervals. Example would be the annual expenditures by single women, by level of income before taxes, selected expenditures. This sample can be inaccurate on the exact amount of dollars that was spent during this period.
Differences is when you “compare averages or percent’s in the sample data to see if there are meaningful differences with percentage difference tests or averages difference test”(Burns, A. & Bush, R. (2012)). An example of this would be the ratio of women’s to men’s earnings by occupation. According to the BLS Spotlight, there was about 81.2 percent ratio for women’s to men’s earnings.
Relationship is the variables that are related to each other. Example for this one would be how women spend their time. This data went from a large range of women’s age from 15 to over 75.
Burns, A. C. & Bush, R. F. (2012). Basic marketing research using Microsoft Excel data analysis (3rd ed.). Upper Saddle River, NJ: Pearson Prentice Hall. Women at Work (2011). U.S. Bureau of Labor Statistics http://www.bls.gov/spotlight/2011/women/pdf/women_bls_spotlight.pdf
Student response 4
In our text the four types of research objectives and data analysis are description, generalization, differences and relationships. Description type of objective summarizes data using percentages, range, averages and metric data. (Burns & Bush 2012) A good example of this type of data is the percentage of women in civilian work force. The labor force summarizes by education. The percentage of women that graduated college, did some college, received a high school diploma or did not complete school. The percentage is base on information received in 2010. Generalization uses findings to the population with hypothesis tests. This type of data could be found in the average of single income. The average single women that makes more than $15,000 before taxes, spend more on transportation. Women that make less than that spend more on shelter and food.
When it comes to the differences type of data it compares averages or percentages in the sample data with difference tests. The data that states the differences in the number of work related injuries between men and women. It states that the percentage of work related inquires are more accidents with men like machines or falls. The percentage is higher with accidents that are violent acts or assaults to women. The relationship analysis uses cross tabulations. An example of this is the way women spend their time during the week. It breaks down age groups are what they if they are well education or at average level. Women 25 years of age and older spend a bigger average of their time out and about, then working or being at home.
Reference
Burns, A. C. & Bush, R. F. (2012). Basic marketing research using Microsoft Excel data analysis (3rd ed.). Upper Saddle River, NJ: Pearson Prentice Hall.
