1. What kinds of biases do you think could be minimized or avoided during the data analysis stage of research ?
· Out range / missing / responses
· Missing data / blank
· Precision and accuracy, data / answers
2. When we collected wonders if he should of treatment in experimental designs which statistical test would be most appropriate to test the treatment effects?
Answer : The most appropriate statistical test to examine the effects of treatment, that is paired t-test test.
3. A tax consultant wonders if he should be more selective about the class of clients he was serves, so as to maximize his income. He usually deals with four categories clients, the very rich, rich, upper middle class and middle class. He has record of each client and each client served, the tax paid by them and how much he has charged them. Since many factors in respect to the clients (number of dependents, business deductibles, etc.) irrespective of the category they belong to, he would like an appropriate analysis to be done to see among the four categories of clientele he should choose to be served in the future.
What kind of analysis should be done the preceding cas and why ?
Answer : Namely multiple linear regression analysis, because we analyze the variation of the various factors which vary widely. With multiple linear regression test and look at the coefficient table 4 we can see among the most influential variables on corporate income by looking at the largest beta value. The largest beta value is the most important and influential on corporate earnings, so should the tax consultant must choose the variable with the largest beta.
4. Below are tables 12A to 12D, summarize the results of data analyses of research conducted in a sales organization that operets 50 different cities of country, with a total sales force of about 500. the number of sales sample for the study was 150:
You are to :
1. Interpret much of the information contained in each table as detailed as possible.
2. Can summarize the results for the CEO of the company.
3. Make recommendations based on the results of your interpretation.
1. a. From table 12A can result in that:
· Sales have averaged 75.1 with a standard deviation of 8.6, minimum 45.2 and maximum sales sales amounted to 97.3
· There are many sellers sebesr average standard deviation of 25 with 6, a minimum of 5 and a maximum of 50
· Population had an average of 5.1 with a standard deviation of 0.8, minimum 2.78 and maximum of 7.12
· Campaigns have an average of 10.3 with a standard deviation of 5.2, a minimum of 6.1 and a maximum of 15.7
b. From table 12B: states that there is a relationship between the variables concerned.
· Correlation coefficient with the seller’s sales amounted to 0.76 means that closeness is strong correlation
· The population correlation coefficient with sales of 0.62 means that closeness is strong correlation
· Correlation coefficient with a sales revenue of 0.56 means keeretan is strong correlation
· Advertising with a correlation coefficient of 0.68 means the closeness of the sales are strong correlation
· With a total population correlation coefficient of 0.06 means that the seller is weak closeness correlation
· Correlation coefficient with the number of ads of 0.16 means that the seller is weak closeness correlation
· Correlation coefficient of income with a population of 0.11 means the closeness of the coefficient of weak
· An ad with a population correlation coefficient of 0.36 means the closeness of the coefficient of weak
· Correlation coefficient with the advertising revenue of 0.23 means the closeness of the correlation is weak.
c. Table 12C
So from the ANOVA table F value of 3.6 & significant F of 0.01. so can we conclude that there was no significant difference between the variance variabel2 who are.
d. 12D table
On the table there are 0.6594, who showed the correlation relationship of the 4 independent variables on the dependent variable (sales) & hubunganny is strong unidirectional nature. Value-me-adjusted coefficient of determination (adjusted R square) of 0.35225 means that 35.225% of sales variable is explained by the independent variable number of sellers, population, income & advertising. 64.775 and the rest is explained by other variables. Standard errors indicate possible errors that may occur if we see dikolom data, we will see that beta is the largest advertising amounted to 0.47 with a significant level of 0.00001 who showed that advertising is the most significant variable affecting.
2. From the table it was concluded that of the variables that most influence the advertising compared with the number of sellers, population, and income, because advertising has the largest beta value.
3. The recommendation that our group is the company must give more attention to advertising by increasing advertising, that could increase sales and corporate earnings.
Kelompok 1 :