Math

Ask a Math Question, Get an Answer ASAP!

Have your own Math question?

4 Math Tutors are Online Now

characters left:

Not a Math Question?

Question

do you have this one in your archives? its for hdur.sav

IN SPSS Download the data set hdur.sav and use SPSS to complete the following calculations.

1) Fit a multiple linear regression model for predicting the hospital duration using age, sex, body temperature, white blood cell counts, antibiotic use, blood culture and service (medication vs. surgery). Using a cut-off value of 0.10 to assess the significance of the predictors. Identify all significant predictors.

2) Assess the model fit for the multiple linear regression model using appropriate statistics and graphics.

3) Assess the assumptions of linear regression in this data using appropriate statistics and graphics.

4) Identify outliers and influential observations using appropriate statistics that can be generated in SPSS. (Hint: You need to do your own research because this is not covered in the textbook or lectures).

I've started it, but I want to make sure I'm on the right track. Hopefully you can help. It's due by midnight tonight.

-Jerel

Submitted: 181 days and 10 hours ago.

Category: Math

Value: $15

Status: AWAITING CUSTOMER ACTION

Level: graduate; Subject: biostatistics

Already Tried:
running analyze regression, analyze descriptive statistics, analyze correlation, analyze compare means one way anova. kind of stumped in deciding how to answer the questions.

Posted by Greg 181 days and 10 hours ago.

Info Request

Alas. No. And my version of SPSS has expired. If you want to send the output and what you've done so far, though, I'd be happy to give it a look and check for errors.

181 days and 10 hours ago.

Reply

1) Fit a multiple linear regression model for predicting the hospital duration using age, sex, body temperature, white blood cell counts, antibiotic use, blood culture and service (medication vs. surgery). Using a cut-off value of 0.10 to assess the significance of the predictors. Identify all significant predictors.

Descriptive Statistics
     N     Minimum     Maximum     Mean     Std. Deviation
Durration of hospitalization     25     3     30     8.60     5.715
age     25     4     82     41.24     20.102
sex     25     1     2     1.56     .507
Body temp     25     96.8     99.5     98.308     .6812
White blood cell count taken     25     3     14     7.84     3.210
Antibiotic use     25     1     2     1.72     .458
Blood culture     25     1     2     1.76     .436
serv     25     1     2     1.64     .490
Valid N (listwise)     25

Descriptives
          N     Mean     Std. Deviation     Std. Error     95% Confidence Interval for Mean     Minimum     Maximum
                              Lower Bound     Upper Bound
Durration of hospitalization     0&n bsp;    16     7.19     3.103     .776     5.53     8.84     3     14
     1     9     11.11  & nbsp;  8.298     2.766     4.73     17.49     3     30
     Total     25     8.60     5.715&nb sp;    1.143     6.24     10.96     3     30
age     0    &nbs p;16     34.69     20.480     5.120     23.77     45.60     4     82
     1     9     52.89& nbsp;    13.688     4.563     42.37     63.41     30     73
     Total     25     41.24    ;  20.102     4.020     32.94     49.54     4     82
sex& nbsp;    0     16     1.44     .512     .128     1.16     1.71     1     2
     1      9     1.78     .441     .147     1.44     2.12     1     2
     Total     25     1.56      .507     .101     1.35     1.77     1     2
Body temp     0     16     98.400     .6335      .1584     98.062     98.738     96.8     99.5
     1   &nbs p; 9     98.144     .7699     .2566     97.553     98.736     97.0     99.5
     Total     25     98.308    & nbsp;.6812     .1362     98.027     98.589     96.8     99.5
White blood cell count taken     0     16     8.75     3.587 &nb sp;   .897     6.84     10.66     3     14
     1 &nbs p;   9     6.22     1.481     .494     5.08     7.36     4     8
     Total     25     7.84     3.2 10     .642     6.51     9.17     3     14
Antibiotic use     0     16     1.69  &n bsp;  .479     .120     1.43     1.94     1     2
     1      9     1.78     .441     .147     1.44     2.12     1     2
     Total     25     1.72  &n bsp;  .458     .092     1.53     1.91     1     2
Blood culture     0     16      ;1.81     .403     .101     1.60     2.03     1     2
     1     9     1.67  & nbsp;  .500     .167     1.28     2.05     1     2
     Total     25     1.76     .436     .087     1.58     1.94     1     2

ANOVA
          Sum of Squares     df     Mean Square     F     Sig.
Durration of hospitalization     Between Groups     88.674     1     88.674     2.933     .100
     Within Groups     695.326     23     30.232
     Total     784.000     24
age     Between Groups     1908.234     1     1908.234     5.634     .026
     Within Groups     7790.326     23     338.710
     Total     9698.560     24
sex     Between Groups     .667     1     .667     2.793     .108
     Within Groups     5.493     23     .239
     Total     6.160     24
Body temp     Between Groups     .376     1     .376     .804     .379
     Within Groups     10.762     23     .468
     Total     11.138     24
White blood cell count taken     Between Groups     36.804     1     36.804     4.020     .057
     Within Groups     210.556     23     9.155
     Total     247.360     24
Antibiotic use     Between Groups     .047     1     .047     .216     .646
     Within Groups     4.993     23     .217
     Total     5.040     24
Blood culture     Between Groups     .122     1     .122     .635     .434
     Within Groups     4.438     23     .193
     Total     4.560     24

2) Assess the model fit for the multiple linear regression model using appropriate statistics and graphics.

3) Assess the assumptions of linear regression in this data using appropriate statistics and graphics.

4) Identify outliers and influential observations using appropriate statistics that can be generated in SPSS. (Hint: You need to do your own research because this is not covered in the textbook or lectures).

I just copied and pasted from what I have so far in my word document. I hope it helps. Tips on how to properly retrieve data on spss using the right menu buttons would be helpful.

_Jerel

Accepted Answer

Great! Here's how I'd answer it:

2) Assess the model fit for the multiple linear regression model using appropriate statistics and graphics.

In this case, you mainly want to look at your predictors to see what ones you should use. Anything with a "sig" (p-value) of less than 0.1, you want to keep. So, you would eliminate everything but age and white blood cell count taken. Those are the only significant predictors of hospital stay. Everything else can be eliminated.

I'm assuming here that your teacher will want you to remove these variables and then run the regression again without them. This is what you would do in real life, but you might get clarification as the instructions do not ask you to do this.

3) Assess the assumptions of linear regression in this data using appropriate statistics and graphics.

For this, you want 2 plots. You want a residual plot. In that plot, you're going to look for patterns. What you want to see is a random scattering of points. If you see a pattern then either you probably have a non-constant variance problem or a non-independent residuals problem.

The second plot is either a Q-Q plot (sometimes called a P-P plot) or a histogram on the residuals. Another of the assumptions is normality. So in the Q-Q plot, you're looking for the points to be in a straight line. If you're looking at a histogram, you'll want to look for that bell-shape.

4) Identify outliers and influential observations using appropriate statistics that can be generated in SPSS. (Hint: You need to do your own research because this is not covered in the textbook or lectures).

I don't know how to make SPSS do this. I'm afraid I'm not too helpful here. I don't often use SPSS.

-----

Let me know if you have questions. I'm happy to clarify any details!

Accepts:
Expert:	Greg
Pos. Feedback:	99.6 %
Answered:	5/25/2009

Graduate Student

I teach math and statistics at a college level as a graduate teaching assistant

181 days and 8 hours ago.

Reply

the data I sent you, was that close to what i'm suppose to be looking for? will i need to run ANOVA, regression, correlation tests?

-Jerel

Posted by Greg 181 days and 8 hours ago.

Answer

You're on the right track. You already have an ANOVA table (that comes with the regression). The correlation also comes with the regression analysis. The output that you have should be sufficient. You just need to figure out how to make SPSS tell you about unusual observations.

Make sense?

Disclaimer: Information in questions, answers, and other posts on this site ("Posts") comes from individual users, not JustAnswer; JustAnswer is not responsible for Posts. Posts are for general information, are not intended to substitute for informed professional advice (medical, legal, veterinary, financial, etc.), or to establish a professional-client relationship. The site and services are provided "as is" with no warranty or representations by JustAnswer regarding the qualifications of Experts. To see what credentials have been verified by a third-party service, please click on the "Verified" symbol in some Experts' profiles. JustAnswer is not intended or designed for EMERGENCY questions which should be directed immediately by telephone or in-person to qualified professionals.

Math

Have your own Math question?

Question

Optional Information

Level: graduate; Subject: biostatistics Already Tried: running analyze regression, analyze descriptive statistics, analyze correlation, analyze compare means one way anova. kind of stumped in deciding how to answer the questions.

Info Request

Alas. No. And my version of SPSS has expired. If you want to send the output and what you've done so far, though, I'd be happy to give it a look and check for errors.

Reply

Accepted Answer

Reply

the data I sent you, was that close to what i'm suppose to be looking for? will i need to run ANOVA, regression, correlation tests? -Jerel

Answer

You're on the right track. You already have an ANOVA table (that comes with the regression). The correlation also comes with the regression analysis. The output that you have should be sufficient. You just need to figure out how to make SPSS tell you about unusual observations. Make sense?

Related Math Questions

Level: graduate; Subject: biostatistics

Already Tried:
running analyze regression, analyze descriptive statistics, analyze correlation, analyze compare means one way anova. kind of stumped in deciding how to answer the questions.

the data I sent you, was that close to what i'm suppose to be looking for? will i need to run ANOVA, regression, correlation tests?

-Jerel

You're on the right track. You already have an ANOVA table (that comes with the regression). The correlation also comes with the regression analysis. The output that you have should be sufficient. You just need to figure out how to make SPSS tell you about unusual observations.

Make sense?