A study was designed to examine student attitudes towards research utilisation in nursing practice (Marshall et al 2007). The Nursing Students' Attitudes and Awareness of Research and Development within Nursing Scale was used to collect data from students enrolled at several Australian schools of nursing
1. Examine the 95% confidence intervals for the mean scores for each university. Based on these, are there any statistically significant differences in mean scores between the universities? 2. The effect sizes (Cohen's d) between the university means were as follows: A and B = 0.02; A and C = 0.17; B and C = 0.14. Using Cohen's conventions, are these small, medium or large effect sizes? 3. Based on your knowledge of statistical power, how confident are you in the interpretation of the following results? What is required to improve the power of this study? (F(df=232) = 0.601; p = 0.549). Post hoc analysis found no difference between all pair combinations of universities p > 0.05. Descriptives
N Mean Std. deviation
Std. error
95% Confidence interval for mean
Minimum Maximum
Lower bound Upper bound University A 87 130.63 13.52 1.45 127.75 133.51 98.00 164.00 University B 80 130.36 13.95 1.56 127.25 133.46 83.00 157.00 University C 66 128.28 14.69 1.81 124.67 131.89 94.00 158.00 Total 233 129.87 13.98 0.92 128.07 131.68 83.00 164.00
(Adapted from: Marshall AP, Fisher MJ, Brammer J, Eustace P, Grech C, Jones B, Kelly M: Assessing psychometric properties of scales: a case study, J Adv Nurs 59(4):398–406, 2007.)
What will be an ideal response?
ANSWERS 1. The 95% CIs for University A and B completely overlap. The 95% CI for University C has smaller upper and lower limits, indicating a slightly lower mean score; however, there is still considerable overlap of University C's 95% CIs with those from the other two universities, hence there is no statistically significant difference between them. 2. The effect sizes between the mean university scores are all small (all < 0.20). A small effect is one that can only be detected statistically, it cannot be directly observed. 3. The small effect size bears out the interpretation of no statistically significant difference between the universities based on the current sample size. In order to determine whether the study is adequately powered, the researchers would need to determine the minimum effect size for clinical significance. If that effect size was, for example, 0.15, to find this effect size statistically significant, the researchers would need to conduct a power analysis to determine the appropriate sample size to detect an effect size of 0.15 with 80% power and alpha = 0.05 (smaller effect sizes require larger sample sizes). They would also need to justify why such a small effect size was clinically relevant. If the clinically relevant effect size was determined to 0.50, then the current sample size is adequate to detect it, should a difference of that magnitude exist between the universities. In this case an effect size of 0.50 did not exist between the universities, so the researchers would conclude no clinically significant differences between them.
You might also like to view...
In an activity log, a nurse records attendance at an education seminar as, "Class on arrhythmia recognition, State College, taught by Sue Smith, RN
" What else would be valuable for the nurse to include? Note: Credit will be given only if all correct choices and no incorrect choices are selected. Standard Text: Select all that apply. 1. The exact title of the seminar or event 2. Which colleagues attended with the nurse 3. Who paid for the course and what it cost 4. The date 5. Any certifications achieved
In which of the following situations would the use of physical restraints most likely be justified?
A) Mr. Y is agitated and aggressive while he is experiencing severe withdrawal from alcohol and is not responding to chemical sedation. B) Mrs. U, who has a diagnosis of dementia, was found wandering outside the hospital and nurses have been unable to redirect her to stay on the unit. C) Mr. I is delirious during the acute stage of his urinary tract infection and is ringing the call bell nearly continuously. D) Mrs. T is frequently entering other patients' rooms and attempting to crawl into others' beds.
A client is receiving chemotherapy with paclitaxel for breast cancer. A priority nursing intervention is to
a. encourage the client to get adequate rest. b. monitor the client's safety. c. teach the client about the drug regimen. d. watch for drug side effects.
A patient is prescribed to take insulin Glargine once per day and regular insulin four times per day. The first dose of regular insulin occurs at the same time of the day as the insulin Glargine dose
Which technique does the nurse use to administer these two drugs? a. Drawing up and injecting the insulin Glargine first, then drawing up and injecting the dose of regular insulin. b. Drawing up and injecting the insulin Glargine first, waiting 1 hour, then drawing up and injecting the dose of regular insulin. c. First drawing up the dose of regular insulin, then drawing up the dose of insulin Glargine in the same syringe, followed by injecting the two insulins together. d. First drawing up the dose of insulin Glargine, then drawing up the dose of regular insulin in the same syringe, followed by injecting the two insulins together.