Biostats SPSS Kaplan-Meier Curve

PART1

Step-by-Step Guide to Assignment 8.1

Problem 1. Data

a. Create a variable for Time-to-event by subtracting the start time (variable = start) from the stop time (variable = stop). Label the new variable “Time”.

Step 1. Open Practice_Week08_dataset.sav. Under Transform, select Compute variable.

Step 2. Type “Time” in the Target Variable box. Select the Time of event or censor [stop] variable and move it to the Numeric Expression box.

Step 3. Stop will appear in the Numeric Expression box. Click on the minus sign on the key pad then click on the Start time for each interval [start] variable and transfer it to the Numeric Expression box. The numeric expression should read “stop – start”. Click OK.

Step 4. Bring up the Data Editor screen in Variable View. Note the new Time variable has 2 decimals. Change this to 0 using the down arrows in the decimal cell for Time.

Step 5. Select Define Variable Properties under Data.

Step 6. In the Define Variable Properties window, select Time in the Variables box and move it to the Variables to Scan box. Click Continue.

Step 7. In the new Define Variable Properties window, type “Time to event/censor” in the Label box. Click OK.

Step 8. Switch to Variable View in the Data Editor Window. The Time variable label should appear as below.

Step 9 Switch to Data View and review the data in Time. Each value should equal the stop value minus the start value.

b. Produce the appropriate descriptive statistics, numerical and graphical, for the following variables:

Step 1. Review each variable to determine whether it is categorical or continuous.

Step 2. Analyze ( Descriptive statistics (Frequencies.

Step 3. Select one variable in Frequencies window and place in Variables box.

For each categorical variable, run frequencies and bar graphs for frequencies:

Step 4. Click on Charts and select Bar Charts and Frequencies. Click Continue. Click OK.

For each continuous variable , run descriptive statistics (mean, standard deviations, minimum, and maximum) and produce histograms for continuous variables.

Step 5. Click on Statistics. Select Mean, Median, and Mode in Central Tendency area. Select Std deviation, Variance, Range, Minimum, Maximum, and SE mean in Dispersion area. Select Skewness and Kurtosis in Distribution area. Click Continue.

Step 6. Click on Charts and select Histogram and Show normal curve on histogram. Click Continue. Click OK.

· Event (categorical)

Tumor
	Frequency	Percent	Valid Percent	Cumulative Percent
Valid	No Tumor	74	42.5	42.5	42.5
	Tumor	100	57.5	57.5	100.0
	Total	174	100.0	100.0

· Interval (categorical)

The order of the time interval
	Frequency	Percent	Valid Percent	Cumulative Percent
Valid	1	80	46.0	46.0	46.0
	2	43	24.7	24.7	70.7
	3	25	14.4	14.4	85.1
	4	18	10.3	10.3	95.4
	5	8	4.6	4.6	100.0
	Total	174	100.0	100.0

· Tx (categorical)

Treatment Status
	Frequency	Percent	Valid Percent	Cumulative Percent
Valid	Placebo	102	58.6	58.6	58.6
	Tx Thiotepa	72	41.4	41.4	100.0
	Total	174	100.0	100.0

· Num (categorical)

Number of Tumors
	Frequency	Percent	Valid Percent	Cumulative Percent
Valid	1	92	52.9	52.9	52.9
	2	27	15.5	15.5	68.4
	3	16	9.2	9.2	77.6
	4	12	6.9	6.9	84.5
	5	12	6.9	6.9	91.4
	6	10	5.7	5.7	97.1
	8	5	2.9	2.9	100.0
	Total	174	100.0	100.0

· Size (categorical)

Size in cm of tumor
	Frequency	Percent	Valid Percent	Cumulative Percent
Valid	1	109	62.6	62.6	62.6
	2	10	5.7	5.7	68.4
	3	36	20.7	20.7	89.1
	4	7	4.0	4.0	93.1
	5	5	2.9	2.9	96.0
	6	5	2.9	2.9	98.9
	7	2	1.1	1.1	100.0
	Total	174	100.0	100.0

· Time (continuous)

Statistics
Time to event/censor
N	Valid	174
	Missing	0
Mean	14.02
Std. Error of Mean	1.026
Median	9.00
Mode	2
Std. Deviation	13.528
Variance	183.011
Skewness	1.232
Std. Error of Skewness	.184
Kurtosis	.694
Std. Error of Kurtosis	.366
Range	59
Minimum	0
Maximum	59