Open patient_data.sav in SPSS. Compute a new composite variable representing how dependent a patient is in performing activities of daily living (ADLs) by summing together the 11 ADL indicators (selfbed, selftran, selfroom, selfcorr, selfon, selfoff, selfdrs, selfeat, selftoi, selfhyg, and selfbath). Compute another new composite measure of pain by multiplying pain frequency (painfreq) by pain intensity (painint). Compute a composite incontinence score by summing the measures of bowel and bladder incontinence (bowcon and bladcon). Add variable labels for the three new variables. Save the data set.
Next, test the following hypotheses.
Print and save your output. Write up your results using the example below as a guide. Be sure to (a) state the null hypothesis being tested, (b) identify and justify the statistical test(s) that are used, (c) present and discuss the values of all relevant statistics, (d) state your alpha level and indicate your decision regarding the null hypothesis, and (e) discuss the implications of your findings for the original hypothesis. Include tables to support your narratives. Turn in your write up and a printout of your output.
Example:
To test the hypothesis that larger nursing facilities have fewer registered nursing hours per patient per day, I tested the null hypothesis that facility size has no linear effect on registered nursing hours. Simple linear regression was used since it is appropriate for determining if a numerical variable (number of beds) has a linear affect on another numerical variable (RN hours per day).
The analysis produced a regression coefficient of .003 which indicates that for every increase of one bed in a facility, the number of RN nursing hours drops by an average of .003 hours. The probability of getting a regression coefficient this large if the null hypothesis of no affect is true is .000. Given that this probability is less than my alpha level of .05 and the effect is negative as was hypothesized, I reject the null hypothesis of no affect and find support for the original hypothesis that larger nursing homes have lower RN nursing hours per patient. The correlation coefficient was .253 (p < .001) and the coefficient of determination was .064, indicating that facility size explains 6.4% of the variation in RN nursing hours.
Table 1. Estimated Linear Regression Coefficient and Standardized Beta Showing Effect of Facility Size on Number of RN Nursing Hours Per Patient Per Day

B 
Std. Error 
Beta 
Constant 
1.031 
.012 

Number of Beds 
.003** 
.000 
.253 
** p < .01.  RSquared = .064. 