PDF document name: LASTNAME_FIRSTNAME_WTAssignment1 Submit via eClass
Note1: WT Assignment 1 is worth a total of 40 marks. Therefore, assignments submitted late were deducted 2 marks per day (40 * 0.05 = 2)
Note2: If you did not use the proper assignment naming convention, 4 marks were deducted.
As noted in my eClass announcement on Nov 5, 2020, you will be deducted marks if you submit a file other than a pdf file and if that file is named incorrectly. For WT Assignment 1, the deduction is 4 marks for incorrect file name. The submission file type has been set up in eClass such that the only accepted file type is PDF.
Other reminders that you should take care to ensure while completing your assignment:
• Questions involving a data file must be answered using SPSS
• HLST 2300 rounding rules apply unless otherwise stated
• Screenshots of any hand-written work and SPSS must be of high resolution and be pasted upright (not sideways) so that they can be easily read and graded
• Answers to questions must directly follow the question asked – do not change the order of the questions
• If you fail to include the SPSS output instructed of you, you will receive zero for any subsequent questions that rely on that output
A researcher has collected data for 158 adult (age ≥ 18 yrs) patients arriving via the Emergency Department (ED) and admitted as an inpatient to Hospital ABC (Excel file: 2300WTassignment1.xls). The data includes the unique patient identifier, sex (female = 1; male = 2), age (years), arrival day of week (DOW) (Sunday = 1; Monday = 2; Tuesday = 3; Wednesday = 4; Thursday = 5; Friday = 6; Saturday = 7), arrival mode (walk-in = 1; ambulance = 2), ED triage level (Resuscitation = 1; Emergency = 2; Urgent = 3), comorbidity level (levels range from 0 – 4 where level 0 represents no significant comorbidity and level 4 represents the group with the largest number of comorbidities), discharge disposition (Discharge Home = 1; Discharged Home with Supports = 2; Transferred to Long-term Care = 3), scores measured at ED arrival, hospital admission and hospital discharge, hospital length of stay (LOS) in days and resource intensity weight (RIW) which is a proxy for hospital resource use.
Before proceeding with any analysis, be sure to:
• Ensure that variables are of correct measure (nominal, ordinal, scale).
• Add labels to all categorical variables. For the variable comorbidity level, add labels Level 0, Level 1,…, Level 4 to the values 0, 1, …, 4.
• Reduce the number of decimal places to 2 for hospital LOS in the variable view (it will likely show 15 decimal places but only requires 2 decimal places).
1. Are comorbidity levels significantly different between males and females?
a) State the specific test you used and why that test was chosen (may require copying and pasting SPSS output table for sample size of appropriate variable). [3 MARKS]
The specific test used was the Mann-Whitney test because:
• we are comparing two groups (males and females);
• the two groups are independent (you can only be either in the male or female group);
• the variable of interest, comorbidity level, is an ordinal variable.
• Did not explain that males and females are the two groups we are comparing. [-1 MARK]
• Did not explain that males and females are independent groups. [-1 MARK]
• Did not explain that comorbidity level is an ordinal variable. [-1 MARK]
b) Copy and paste the relevant SPSS output table(s) used in reporting results. [2 MARKS]
• Did not copy and paste the Mann-Whitney Ranks and Test Statistics table or values in the tables are different than those shown. [-1 MARK]
• Did not copy and paste the Percentiles table or values in the tables are different than those shown (or descriptives table with correct IQRs for comorbidity level for each group). [-1 MARK]
c) Report the results, including showing your calculations for effect size if point estimate not included in SPSS output. [4 MARKS]
A Mann-Whitney test was used to test the hypothesis that the distribution of comorbidity levels between females and males was equal because comorbidity level was measured on an ordinal scale. Comorbidity levels in females (Weighted Average Mdn = 1.00, IQR [0.00, 2.00]) were not statistically significantly different than males (Weighted Average Mdn = 0.50, IQR [0.00, 2.00]), U = 2638.00, z = -
−1.143
1.14, p = .255, 𝑟 = = -.09.
√158
• Note, if reported results without producing the corresponding tables (part b) where these results came from, no marks are earned for part c.
• Did not report the median and IQR comorbidity levels for both groups correctly. [-1 MARK]
• Did not report U and z statistics correctly. [-1 MARK]
• Did not report p-value correctly (since we have a two-tailed hypothesis, we will report exact significant (two-tailed)) and state that result was not statistically significant. [-1 MARK]
• Did not show calculation and report r correctly. [-1 MARK]
• Rounding errors. [-0.5 MARKS]
2. Are scores measured at hospital admission significantly different than those at hospital discharge?
a) State the specific test you used and why that test was chosen (may require copying and pasting SPSS output table for sample size of appropriate variable). [4 MARKS]
The specific test used was the paired t-test because:
• we are comparing two groups (admission and discharge);
• the two groups are repeated (same group of patients are being tested in both conditions – admission and discharge);
• the variable of interest, score, is a scalar variable. Since the sample size is large (n 30): nscore at admission = nscore at discharge = 158, the CLT states that parametric test is robust even if the assumption of normality is not met.
• Did not explain that admission and discharge are the two groups we are comparing. [-1 MARK]
• Did not explain that admission and discharge are repeated groups. [-1 MARK]
• Did not explain that score is a scale variable. [-1 MARK]
• Did not include a table that indicates the number of observations for both admission and discharge scores. [-1 MARK]
b) Copy and paste the relevant SPSS output table(s) used in reporting results. [2 MARKS]
• Did not copy and paste the Paired Samples Test table or values in the table are different than those shown. Note that if you chose to enter score_discharge – score_admission, this is fine; the mean difference, CI and t values will have the same magnitude but the opposite signs. [-1 MARK]
• Did not copy and paste the Paired Samples Effect Sizes table or values in the table are different than those shown. Note that if you chose to enter score_discharge – score_admission, this is fine; the d value (point estimate) will have the same magnitude but the opposite sign. [-1 MARK]
c) Report the results, including showing your calculations for effect size if point estimate not included in SPSS output. [5 MARKS]
On average, admission scores (M = 52.07, SE = 1.15) were different than discharge scores (M = 77.16, SE = 1.21). The mean difference, -25.08, 95% CI [-28.18, -21.99] was statistically significant t(157) = -16.00, p <.001; and represents a very large effect, d = -1.69.
• Note, if reported results without producing the corresponding tables (part b) where these results came from, no marks are earned for part c.
• Note that if you chose to enter score_discharge – score_admission, this is fine; the mean difference, CI, t and d values will have the same magnitude but the opposite signs.
• Did not report the mean and standard error for the mean scores for both groups correctly. [-1 MARK]
• Did not report the mean difference and 95% CI of the mean difference correctly. [-1 MARK]
• Did not report t-value and degrees of freedom correctly. [-1 MARK]
• Did not report p-value correctly and state that result was statistically significant. [-1 MARK]
• Did not report d correctly. [-1 MARK]
• Rounding errors. [-0.5 MARKS]
3. Among Friday ED arrivals, are scores measured at hospital admission significantly smaller than those at ED arrival?
a) State the specific test you used and why that test was chosen (may require copying and pasting SPSS output table for sample size of appropriate variable). [4 MARKS]
The specific test used was the Wilcoxon Signed-Rank test because:
• we are comparing two groups (arrival and admission);
• the two groups are repeated (same group of patients are being tested in both conditions – arrival and admission);
• score is a scalar variable but the sample size for Friday arrivals is small (n < 30): nscore at arrival = nscore at admission = 23, and we’re not told if scores are normally distributed in the population
• Did not explain that arrival and admission are the two groups we are comparing. [-1 MARK]
• Did not explain that arrival and admission are repeated groups. [-1 MARK]
• Did not explain that score is a scale variable. [-1 MARK]
• Did not include a table that indicates the number of observations for both arrival and admission scores, specific to Friday arrivals. [-1 MARK]
• Did not add labels to arrival day of the week (Friday is indicated as a value ‘6’ rather than its appropriate label ‘Friday’). [-0.5 MARKS]
b) Copy and paste the relevant SPSS output table(s) used in reporting results. [2 MARKS]
• Did not copy and paste a descriptives statistics table indicating Friday median and IQR for arrival and admission scores or values in the table are different than those shown. [-1 MARK]
• Did not copy and paste the Ranks and Test Statistics tables or values in the table are different than those shown. Note that if you chose to enter score_arrival – score_admission, this is fine; the row labelled Negative Ranks would have the values in the Positive Ranks row and vice versa. [-1 MARK]
c) Report the results, including showing your calculations for effect size if point estimate not included in SPSS output. [4 MARKS]
The Wilcoxon Signed-Rank test was used to determine whether, among Friday arrivals, scores at admission were significantly smaller than scores at arrival since we had small samples. While scores at admission (Mdn = 47.93, IQR [37.19, 65.29]) were smaller than scores at arrival (Mdn = 52.29, IQR [37.61, 66.09]), the result was not statistically significant, T = 87.00 (or 189.00 if score_arrival –
−1.551
score_admission), z = -1.55, p = .062, r = = -.23.
√46
• Note, if reported results without producing the corresponding tables (part b) where these results came from, no marks are earned for part c.
• Did not report the median and IQR scores for both groups correctly. [-1 MARK]
• Did not report T and z statistics correctly. [-1 MARK]
• Did not report p-value correctly (since we had a one-tailed hypothesis, we will report exact significant (one-tailed)) and state that result was not statistically significant. [-1 MARK]
• Did not show calculation and report r correctly. [-1 MARK]
• Rounding errors. [-0.5 MARKS]
4. Among ED patients with triage level Urgent, is hospital LOS significantly longer than 8 days?
a) State the specific test you used and why that test was chosen (may require copying and pasting SPSS output table for sample size of appropriate variable). [3 MARKS]
The specific test used was the one-sample t-test because:
• we are comparing one group’s LOS (group = ED patients with triage level Urgent) against a specified constant;
• hospital LOS is a scalar variable and since our sample size is large (n 30): nurgent_LOS = 45, the CLT states that parametric test is robust even if the assumption of normality is not met.
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
