Lab 3 worksheet Descriptive Data JASP.docx PDF

Preview

Summary

Download Lab 3 worksheet Descriptive Data JASP.docx PDF

Description

Name: Thalia Orlando

PSYCH248 Lab 3 Worksheet/JASP

Put your answers in blue. Part 1. Let’s see how much you have already learned about entering data in JASP and using it to understand data. Here are data that show the number of words recalled in a learning experiment for two groups of participants: first-grade children and college students. Enter the data into Excel and open it in JASP. It is always a good idea to check for accuracy before you begin generating results. For example is there missing data, does it look like a data point might be wrong etc.…Think about how to organize it (whether in rows or in columns) — you need to compare two groups. First grade students: 26, 21, 14, 12, 29, 17, 7, 28, 14, 13, 8, 27, 11, 20, 17, 18, 6, 6, 7, 16 College students: 28, 23, 25, 18, 20, 24, 23, 21, 16, 15, 27, 21, 19, 26, 24, 25, 21, 30, 21, 19, 32, 24, 31, 27, 27, 26, 23, 17, 31 1. What are the dependent and independent variables for this study? INDEPENDENT VARIABLES: FIRST GRADE AND COLLEGE STUDENTS DEPENDENT VARIABLES: NUMBER OF WORDS 2. Construct a histogram (distribution plot) with normal curve for each distribution. What shape would you call each curve? WORDS

1

2

WORDS College students distribution is more normal than 1st grade students. 1st grade is fairly symmetrical.

2a. Create a box plot showing both groups of college students and children and paste it below. Which graph (histogram or boxplot) do you prefer for presenting the data and why?

Boxplots WORDS

3. Report the mean, median, mode, standard deviation, variance, interquartile range and skewness for both children and college student’s data. Compare the performance of the groups on the recall test, and discuss the shape of each group. Remember that to describe data you should address three aspects of the distribution: 1) form (distribution shape), 2) center (average), and 3) spread (sd). THE COLLEGE STUDENTS SKEWNESS VALUE IS SMALLER, THEY ARE MORE NORMALLY DISTRIBUTED. Descriptive Statistics

WORDS 1 Valid Missing Mean Median Mode Std. Deviation Skewness Std. Error of Skewness Minimum Maximum

ᵃ

20 0 15.850 15.000 6.000 7.485 0.381 0.512 6.000 29.000

2 29 0 23.586 24.000 21.000 4.555 0.031 0.434 15.000 32.000

ᵃ More than one mode exists, only the first is reported

4. How much do the two groups overlap (the box plot might be most helpful for answering this)? How do the sharpest first grade students compare with college students?

There is some overlap, our highest whiskers are close but the means are still kind of far from one another. The first grade students are more “variable” because the box plot is higher which means it has more different word values. The sharpest first graders is pretty similar to the sharpest college students. 5. If you wanted to obtain one mean for the entire data set, which function would be better to use on the data, the mean or the weighted mean?

The weighted mean because we have two different group sizes, we have more college students which means we would expect more. 5a.What are the results for the weighted mean and which group has more “weight” in this result and why? 317+684/20+29=1,001/49=20.4285714 COLLEGE STUDENTS HAS MORE WEIGHT BECAUSE THERE’S MORE DATA POINTS/VALUES. 6. Find the 10th, and 90th percentile for both data. In the statistics area click the “percentiles” box and enter the percentage you want to find out about. At which percentile is there a larger discrepancy in the number of words recalled between the two groups?

10TH PERCENTILE 1ST GRADE: 6.900, COLLEGE STUDENTS: 17.800 90TH PERCENTILE 1ST GRADE: 27.100, COLLEGE STUDENTS:30.200 THE 10TH PERCENTILE HAS A BIGGER DIFFERENCE. 6a. What does it mean to be in the 90th percentile for this data?

IT MEANS IT FALLS ABOVE 90 TH ABOVE OTHERS AND THEY HAVE SCORED 90% HIGHER THAN OTHER PARTICIPANTS. Part 2: A researcher wants to determine the nutritional value of school lunches at a local school. As part of this study, the researcher reports the number of calories in each meal and summarizes the data in two ways. The first way is performing descriptive statistics and looking at the general caloric values of the lunches. The second way creating three separate groups to see how many low, medium, to high lunches are being offered at the school. Histogram: The researcher reports the calories per meal. The data for 30 school lunches are given below. Since calories are a quantitative variable, the data can be summarized using a histogram. Enter the following data into excel and bring into JASP. Calories per meal: 350, 330, 880, 500, 490, 390, 540, 640, 900, 435, 695, 440, 730, 660, 620, 680,860, 920, 705, 675, 600, 880, 510, 720, 910, 500, 665, 805, 565, 475 Construct a histogram (distribution plot) and normal curve (display density) on these data in JASP and copy and paste it below. CALORIES PER MEAL

1. Does the curve look fairly normal? YES

The researcher also summarizes the frequency of meals that are low calorie (less than 450 calories), moderate calorie (between 450 and 650 calories), and high calorie (more than 650 calories). The data given above should be used to create a categorical (ordinal) variable for calories. Go back to the original excel file and highlight (clicking the letter on top) and filter the data from lowest to highest by clicking the A to Z button and select “ascending”. In the column to the right on the data you can code all data less tan 450 as 1, between 450 and 605 calories as 2, and data more than 650 calories as 3. Bring the CSV file into JASP and rename the dummy code to 1 as low, 2 as moderate, and 3 as high by clicking on the variable name at the top of the data. Click on each number and give appropriate name. Click on the descriptives function and move the calorie data into the variable box and split it by your category variable. Select the frequency box to see the frequencies of each meal given. 2. Entering the scale data you made in descriptives-create a frequency table. Which caloric meal is offered the most to the students? HIGH 3. Create a box plot of the categorical data by going into plots choosing Boxplots and then Boxplot element, and paste it below. Looking at the boxplots, which data set is the most normally distributed and what aspects of the plot did you use to determine this? ____

Boxplots CALORIES PER MEAL

THE MEAL WITH THE LOW CALORIES PER MEAL IS MORE NORMALLY DISTRIBUTED BECAUSE THE MEDIAN IS IN THE MIDDLE OF THE BOX AND THE OUTLIERS AREN’T AS FAR A PART. _______________________________________________________________________________________ 4. Now obtain z-scores for the calorie data in excel by taking your sorted calorie data and selecting it and pasting it on a new sheet within the file. Obtain the mean and the standard deviation for the data set. In the cell directly under the data insert an equal sign and type “average” and then select the calorie data. Under the average obtain the standard deviation for the data by inserting “=standev” and selecting the data. Next you can enter the z-score equation in the cell next to the first calorie datum point. If your first data point is in cell A2, the equation will be: “=STANDARDIZE(A2,A$32,A$33)”. The $ sign makes sure the reference cell does not move. After obtaining the z-score for the first data point, click on the first z-score and drag down box to obtain the z-scores for the rest of the data. a. How many data points have an absolute z-score of 1.5 or higher? 5 b. What percentage of the meals have a z-score of 1.5 or greater? 10% c. What caloric amount does this z-score (z =1.5) align with? 880...