Order Times of Burger King and Mcdonald S

Order Times of Burger King and McDonald’s

Initially as we began this project, we decided that we wanted to do an experiment that could be of use to people with their everyday lives. Our first idea was to see the effect that chewing gum has on stress levels of people. We began to set up this experiment, but as we got further and further into the process we realized that it would be difficult to consistently get accurate stress levels. So, we chose to drop that idea and find the difference in order times at McDonalds and Burger King fast food restaurants and their drive through and walk in differences as well. The purpose of this experiment is to see which fast food restaurant is fastest. We choose to do this test by going to different Burger King and McDonalds locations and timing customer’s order times.

The populations we will be studying is the customers at Burger King and McDonalds fast food restaurants. We sampled two Burger Kings and two McDonalds restaurants with similar locations. The locations of the Burger Kings we sampled were 7800 Peach Street and 2872 West 26th Street. The locations of the McDonalds restaurants we sampled were 7200 Peach Street and 2643 West 26th Street.

The intent of this experiment is to see which fast food restaurant serves their customers the fastest, and where the customers came through the restaurant, the drive through or walking in. The time was started once the customer finished placing their order and the time was stopped once they received their food. Our study is a two sample T-test. The reason it falls into this category is due to the fact that we are comparing means. In order to create randomization between order sizes we had to obtain a very large sample size to help minimize the effect of this lurking variable. The size of the orders could be a critical issue; however, with a large number of samples this should be fairly random. Time of day and day of the week could be a lurking variable, so we starting taking data from each location at exactly noon on a work day. Also, location and number of workers on duty are variables that could skew our results, so in order to minimize these we choose locations that have Burger King and McDonalds restaurants right down the street from each other and started sampling at the same time of day, noon, so the number of people on shift should be very similar. We sampled every customer that went ordered food at each of our sampling locations until we had obtained twenty-five customers. We used stop watches to keep track of the customer’s order times, 8 stop watches to be exact. We were lucky enough to borrow stop watches from Mr. Drabina as well as Mr. Jensen and we thank them for their help. It is hard to apply the result we got to all of the Burger King and McDonalds fast food restaurants in the city, but we can apply this to what street has the faster fast food, West 26th Street or Peach Street.

We went to each of the restaurants during or class time which happened to be perfect for us to arrive at our sampling location before noon to sample. One of us sat outside in the car and took times of the drive through while the other sat inside the restaurant and took data for the walk in customers. There was a chance for bias to be present, but we took the times of every customer starting at noon. The only legitimate source of bias in this experiment could be experimenter bias. We minimized this by not skewing any results, and writing down the exact times that we got on each stop watch, and rounding down when necessary. No randomization of customers was used in order to obtain a large sample size, but this should not be too big of an issue due to the consistency we had with every location we sampled. We wanted to make our sampling techniques the same from location to location to help rid of lurking variables, and we did make our sampling techniques the same.

Sampling Method:

We drove to each location on a work day of the week, and took the order times of every customer that ordered food starting at noon till we obtained twenty-five samples. The order time began once the order was finished being placed, and ended once the customer received their food.

Statistical Procedures and Calculations:

2 sample T-Test of Combined drive through and Walk In from Burger King and McDonalds

Assumptions:

Simple Random Sample- No, this would have taken too much time to decide what cars to sample, but since this is consistent between locations it should not be too big of an issue.

Independent- Yes, usually the same customers will not go to Burger King and McDonalds at the same time on the same type of work day, but there is a little reason to question.

10% Condition- Yes, this sample is not even close to the 10% of customers these restaurants receive in a day.

Nearly Normal Condition- Yes, the samples of data of all BK times was a little skewed to the right, but the large sample size that is over 30 making this condition okay. Also, the data of MCD was normal and also had a large sample size that is over 30 making this condition met.

HO : BK – MCD = 0

HA : BK – MCD ≠ 0

P-Value = 4.135978*10^-10

t = -6.624887132

df = 174.8260089

Mean of BK = 105.88 seconds

Mean of MCD = 163.5 seconds

STD of BK = 49.0434316 seconds

STD of MCD = 71.8289916 seconds

n of BK = 100 customers

n of MCD = 100 customers

Variances not equal

Assuming there is no difference in order times of BK and MCD, there is about a 0% chance of getting results like we did. This is strong evidence that the order times of these two restaurants are not equal and that BK has overall faster order times.

2 sample T-Test and T-Interval of Drive through from Burger King and McDonalds

Assumptions:

Simple Random Sample- No, this would have taken too much time to decide what cars to sample, but since this is consistent between locations it should not be too big of an issue.

Independent- Yes, usually the same customers will not go to Burger King and McDonalds at the same time on the same type of work day, but there is a little reason to question.

10% Condition- Yes, this sample is not even close to the 10% of customers these restaurants receive in a day through the drive through.

Nearly Normal Condition- Yes, the samples of data of all BK times was a very skewed to the right, but the large sample size that is over 30 making this condition okay. Also, the data of MCD was a little skewed to the right, it also had a large sample size that is over 30 making this condition met.

HO : BK – MCD = 0

HA : BK – MCD ≠ 0

P-Value = 7.0621607*10^-9

t = -6.44165284

df = 83.85393563

Mean of BK = 92.16 seconds

Mean of MCD = 179.58 seconds

STD of BK = 52.0883707 seconds

STD of MCD = 80.5944977 seconds

n of BK = 50 customers

n of MCD = 50 customers

Variances not equal

Assuming there is no difference in order times of BK and MCD drive through, there is about a 0% chance of getting results like we did. This is strong evidence that the order times of these two restaurants drive through’s are not equal and that BK has overall faster order times through the drive through.

Confidence Interval: .95

T-Interval: (-114.4 to -60.43)

Using these methods, 95% of samples will produce an interval that contains the mean difference of drive through order times at Burger King and McDonalds. Therefore we reject the null hypothesis that Burger King and McDonalds had the same order times because zero is not contained in the interval.

2 sample T-Test of Walk In from Burger King and McDonalds

Assumptions:

Simple Random Sample- No, this would have taken too much time to decide what cars to sample, but since this is consistent between locations it should not be too big of an issue.

Independent- Yes, usually the same customers will not go to Burger King and McDonalds at the same time on the same type of work day, but there is a little reason to question.

10% Condition- Yes, this sample is not even close to the 10% of customers these restaurants receive in a day and walk in.

Nearly Normal Condition- Yes, the samples of data of all BK times was normally distributed, and the large sample size that is over 30 making this condition okay. Also, the data of MCD was fairly normal also, and it also had a large sample size that is over 30 making this condition met.

HO : BK – MCD = 0

HA : BK – MCD ≠ 0

P-Value = .0166807691

t = -2.440933296

df = 86.80120586

Mean of BK = 123 seconds

Mean of MCD = 147.42 seconds

STD of BK = 40.0428342 seconds

STD of MCD = 58.3175998 seconds

n of BK = 50 customers

n of MCD = 50 customers

Variances not equal

Assuming there is no difference in order times of BK and MCD walk in, there is about a 2% chance of getting results like we did. This is strong evidence that the order times of these two restaurants walk in time’s are not equal and that BK has overall faster order times inside the restaurant.

Conclusion:

The conclusions that we choose to make with our data all show that we must reject our null hypothesis that Burger King and McDonalds order times overall, through the drive through, and through the walk in are equal. With each test we found that Burger King had the faster order times, but the walk in times were not as extreme as the combined and drive through results, but was still strong evidence that Burger King was faster. The 2 sample t-interval solidified that Burger King has faster order times than McDonalds because zero was not contained in our interval of -114.4 to -60.43 seconds.

If we were to do this experiment again, there are a few things that we would like to perhaps do differently. First of all, we would like to sample more restaurants in the Erie area. This would make our population much larger, and allow for us to apply the result to all Burger Kings and McDonalds in Erie. Also, we would like to get data from the same location more than once. This will account for a possible lurking variable of the lack of workers. Another thing we would like to do differently is to randomly select the customers that we timed. This would give randomization to the experiment, and help minimize the effect of a hold up in the line due to various uncontrollable reasons, such as people who do not leave once they get their food and a back up of customers.

Overall, we enjoyed taking time out of our busy lives to complete this experiment. One reason is simply because our data can help save us time in the future. We believe that our experiment shows proof that Burger King is faster than McDonalds most of the time. We realize that there is a possibility for a large amount of lurking variables to affect our data. We feel that our consistency minimized the effect on our results, and in the future we would like to get even more data to take steps towards making these lurking variables a non factor. In the end, we feel that our experiment was a success and that our results can help save the valuable time of the average fast food consumer.

Appendix: