Stock automobiles containing dummies in the driver seats crashed into a wall at 35 miles per hour. National Transportation Safety Board officials collected info on how the crash affected the dummies. One variable scored the extent of chest injuries sustained by the dummies (hihger score=more extensive injury)
We were given a list of 51 different cases noting the make of the car, the model of the car, the head injury, the chest injury, and the wt size (for example the first one was chevrolet (make), Beretta (model) 343 (head injury) 37(chest injury) 2671 comp (size)
Perform a one way ANOVA using spss to test for different chest-injury schores in the different makes of cars. Is there a significant effect )at the 5% level) If so perform multiple comparisons using spss using the bonferroni method. What is your overall conclusion?
First I put it all in spss but I have a couple of questions...
1) for the size do i literallywrite 2671 comp or just plain 2671 or does it even matter?
2) I tried getting a one way anova table...im not sure exactly what table im looking for I thought i was just looking for a table that contains the sum of squaresbetween,within, and total ; DFbetween, within total ; mean squarebetween,within,total ; F-ration between ; significance between or do they not give this information to you?
I tried to press "analyze"-->"compare means"-->one way anova...a box came up and said what i wanted to put as the independent variable and as the factor. The only two choice it gave me to put in these boxes was chest injury and head injury but i know that the independent variable is chest injury but the "factor" i thought was model but they did not give me an option of putting it in there (I think this is because when i click "variable view" at the bottom the "make", "model", and "wt size" are considered "string" (no idea what this means), and nominal (I guess wt size is considered that becuase i typed in "comp" next to the number.) Chest injury and head injury are both considered "numeric" and "scaled"...so i am not sure how to get Model into the box that says "factor"....i tried changing "model" in the variable view to scale, but when i went back to the list all the models had question marks next to it...I tried changing "model" to numeric but the same thing happened....then i did both and again the question marks popped up
So i found another way to try and produce an Anova table by clicking analyze--->general linear model-->univariate now I got my dependent variable as "chest injury" and factor as "model" which was good then i pressed options and clicked all the boxes (homogenity test, parameter estimate, observed power, lack of fit, descriptive statistic etc...) hoping that one of them would be the ANOVA but none of them were...
If anyone can help me-id really appreciate it!!
Well, isn't this crash test dummy business a real smash hit? Let me try to help you navigate through this twisted wreckage of confusion and get you back on track:
1) As for the weight size, you can simply input the number 2671. The "comp" you mentioned is probably just an abbreviation for "completed" or "completed cases" indicating that the weight size was measured completely. So don't worry, there's no need to complicate things with extra words.
2) Ah, the elusive ANOVA table you seek! In SPSS, it might not be available in the straightforward manner you described. However, fear not, for there is another way to obtain it. Once you have your data inputted correctly, go to "Analyze", then "General Linear Model", and choose "Univariate". Here, you can select "Chest Injury" as your dependent variable and "Model" as your factor. Click on "Options" and you should see the magical "Tests of Between-Subjects Effects" table, which contains the sum of squares, degrees of freedom, mean squares, F-ratios, and significances you desire. Huzzah!
Now, as for the multiple comparisons using the Bonferroni method, you can achieve that by going to "Post Hoc..." in the "General Linear Model" window. Choose "Model" as the factor, click on "Compare Main Effects", and select the Bonferroni method from the drop-down menu. This will provide you with the p-values for the pair-wise comparisons between car models in terms of chest injuries.
Once you have all the results, take a step back, analyze the data, and draw your overall conclusion. Are some car makes inflicting more damage on these poor dummies? Or is there no significant difference worth dinging their reputation?
1) For the size variable, you can simply write "2671" without "comp". The additional information "comp" might be indicating the unit or some other specific context, but for the purpose of your analysis, the numerical value "2671" should be sufficient.
2) To perform a one-way ANOVA in SPSS and obtain the ANOVA table, you can follow these steps:
- Click on "Analyze" from the top menu.
- Choose "General Linear Model" and then "Univariate".
- In the "Dependent Variable" box, enter "chest injury".
- In the "Fixed Factors" box, enter "make" (assuming "make" refers to the different car makes).
- Click on the "Options" button and select the relevant options you want to include (e.g., descriptive statistics, homogeneity tests, etc.).
- Click "OK" to run the analysis.
After performing these steps, SPSS will generate the ANOVA table, which will include the sum of squares between, within, and total; the degrees of freedom between, within, and total; the mean squares between and within; the F-ratio; and the significance value.
Regarding the issue with "model" being a nominal variable, you can convert it into a numerical variable for the purpose of running the ANOVA. To do this, go to the "Variable View" in SPSS, select the "model" variable, and change the "Measurement Level" from "Nominal" to "Scale". However, note that this conversion may not be appropriate if the "model" variable represents different categories rather than a continuous numerical scale.
Overall, it's important to ensure that the variables you are using in the analysis are correctly assigned the appropriate measurement levels (e.g., scale, nominal, ordinal) to ensure meaningful interpretation of results.
Sure, I can try to help you with your questions.
1) For the size variable, you should enter the numerical value "2671" in SPSS. The "comp" part seems to be a label or abbreviation, so you can ignore that.
2) To obtain the ANOVA table, you can follow these steps in SPSS:
a) Go to "Analyze" in the top menu, then select "General Linear Model" and choose "Univariate".
b) In the "Dependent Variable" box, select "chest injury".
c) In the "Fixed Factors" area, move the "model" variable to the "Model" box.
d) Click on the "Options" button and make sure the "Descriptive statistics" box is checked.
e) Click on "Continue" and then click on "OK".
After these steps, SPSS should generate the ANOVA table showing the sum of squares (SS), degree of freedom (df), mean squares (MS), F-ratio, and significance level for the between-groups effect (model).
Regarding your concern about the "make", "model", and "wt size" variables being classified as "string" and "nominal", it is expected since these variables contain categorical information. However, for the ANOVA analysis, you are interested in the relationship between the "chest injury" variable (numeric and scaled) and the different car models (categorical). So, you don't need to change the data type of the "model" variable.
Once you have the ANOVA results, you can perform multiple comparisons using the Bonferroni method. To do this in SPSS:
a) Go to "Analyze" in the top menu, then select "Compare Means" and choose "One-Way ANOVA".
b) In the "Variables" box, select "chest injury".
c) In the "Factor" field, select "model".
d) Click on the "Post Hoc" button and choose "Bonferroni" as the method.
e) Click on "Continue" and then click on "OK".
SPSS will provide you with the post hoc comparison results using the Bonferroni correction.
Regarding your overall conclusion, you will need to interpret the results of the ANOVA and the multiple comparisons to determine if there is a significant effect of the car make on chest injury scores. If there is a significant effect, the multiple comparisons can help you identify which car makes have significantly different chest injury scores from each other.
I hope this helps! Let me know if you have any more questions.