Question

Please help.I would normally work on this my own but I really don't understand these two questions and they're the only ones that I'm stuck on

1. The owner of a large company is conducting a survey about job satisfaction, including questions about salary, hours, stress, and other conditions. There are about 800 employees in the company, including 100 executive positions, 650 middle-management positions, and 50 custodial positions. The owner wants to include about 10% of his employees in the survey sample.

Identify a sampling method that might produce a biased sample. Explain why using

Answer 1

2. The owner of a large company is conducting a survey about job satisfaction, including questions about salary, hours, stress, and other conditions. There are about 800 employees in the company, including 100 executive positions, 650 middle-management positions, and 50 custodial positions. The owner wants to include about 10% of his employees in the survey sample.

Identify a sampling method that would lead to a representative sample for the survey. Explain why it would be a good choice and give details about the process.

Answer 2

Do you understand what these two questions mean when they talk about a biased sample and a representative sample?

Answer 3

No, not really. :/

Answer 4

A sample should represent what the actual population is like. You could just ask ALL employees, but that would take a long time... so you ask 10% of the employees to save time, but you want a 10% sample that matches as well as you can what the full set of employees thinks.

A biased sample is one that is not a realistic -- it gives results that are biased, meaning that the results say one thing, but the full population actually says something different, and the problem is caused by a bad choice in the 10% of people.

Answer 5

So, silly example, a class of 19 boys and 1 girl... teacher wants to take a sample of class interest in a party theme... teacher asks 10%, or 2 kids, what they want. Boy wants ARMY PARTY, girl wants FLOWER PARTY, and the teacher concludes, incorrectly, that 50% of the class wants each type of party.

The problem was that her 10% sample gave way too much weight to the girl's interests, giving her as much of a vote as 10 boys.

Answer 6

So a biased sample in your problem might be one that asked the same number of people from each group, even though some groups have many more people than others.

Answer 7

10% of 800 employees is 80, and you have 3 groups, so maybe you choose 26 people from one group and 27 from the other two groups. But this method will BIAS your sample in favor of the custodians, since they have so many fewer employees than the other two groups... just like the teacher biased her sample in favor of the girl, who got 50% of the vote even though she was 5% of the class.

Answer 8

Ohhh, okay I get it now.

Answer 9

The idea is to fair... although it might seem unfair depending on your point of view.

An unbiased sample on problem 1 might take 10% of each group. However, the custodians might complain that only 5 of them got surveyed vs.10 executives and 65 mid-management. But that would be an "unbiased" or "representative" sample...

Answer 10

oh, guess I just addressed problem 2... :)

I have to run... hope this helped!! :)

Answer 11

:) Okay, thanks for your help! I really appreciate it!