Skip to content
Textbook Guides

Solution Manual For Statistics for Nursing Research: A Workbook for Evidence-Based Practice, 3rd Edition

Struggling with textbook exercises? Solution Manual For Statistics for Nursing Research: A Workbook for Evidence-Based Practice, 3rd Edition breaks down solutions in a way that�s easy to understand.

Preview 16 of 88 pages

See the first 16 pages exactly as they appear in the document.

  1. Page 1
    Read page 1 as text

    AG 1-1 Answer Guidelines for Questions to Be Graded The questions are in bold followed by answers. 1. In Table 1, identify the level of measurement for the current therapy variable. Provide a rationale for your answer. Answer: The current therapy variable was measured at the nominal level. These drug categories were probably developed to be exhaustive for this study and included the categories of drugs the subjects were receiving. However, the categories are not exclusive, since patients are usually on more than one category of these drugs to manage their health problems. The current therapies are not measured at the ordinal level because they cannot be rank ordered, since no drug category can be considered more or less beneficial than another drug category (see Figure 1-1 ; Grove & Gray, 2019 ). 2. What is the mode for the current therapy variable in this study? Provide a rationale for your answer. Answer: The mode for current therapy was  blocker. A total of 100 (94%) of the cardiac patients were receiving this category of drug, which was the most common prescribed drug for this sample. 3. What statistics were conducted to describe the BMI of the cardiac patients in this sample? Discuss whether these analysis techniques were appropriate or inappropriate. Answer: BMI was described with a mean and standard deviation ( SD ). BMI measurement resulted in ratio-level data with continuous values and an absolute zero (Stone & Frazier, 2017). Ratio- level data should be analyzed with parametric statistics such as the mean and SD ( Grove & Gray, 2017 ; Knapp, 2017 ). 4. Researchers used the following item to measure registered nurses’ (RNs) income in a study: What category identifies your current income as an RN? a. Less than $50,000 b. $50,000 to 59,999 c. $60,000 to 69,999 d. $70,000 to 80,000 e. $80,000 or greater What level of measurement is this income variable? Does the income variable follow the rules outlined in Figure 1-1? Provide a rationale for your answer. Answer: In this example, the income variable is measured at the ordinal level. The income catego- ries are exhaustive, ranging from less than $50,000 to greater than $80,000. The two open-ended Identifying Levels of Measurement: Nominal, Ordinal, Interval, and Ratio EXERCISE 1

  2. Page 2
  3. Page 3
    Read page 3 as text

    AG 1-2 Answer Guidelines for Questions to Be Graded categories ensure that all salary levels are covered. The categories are not exclusive, since catego- ries (d) and (e) include an $80,000 salary, so study participants making $80,000 might mark either (d) or (e) or both categories, resulting in erroneous data. Category (e) could be changed to greater than $80,000, making the categories exclusive. The categories can be rank ordered from the lowest salary to the highest salary, which is consistent with ordinal data (Grove & Gray, 2019 ; Waltz et al., 2017 ). 5. What level of measurement is the CDS score? Provide a rationale for your answer. Answer: The CDS score is at the interval level of measurement. The CDS is a 26-item Likert scale developed to measure depression in cardiac patients. Study participants rated their symp- toms on a scale of 1 to 7, with higher numbers indicating increased severity in the depression symptoms. The total scores for each subject obtained from this multi-item scale are considered to be at the interval level of measurement ( Gray et al., 2017 ; Waltz et al., 2017 ). 6. Were nonparametric or parametric analysis techniques used to analyze the CDS scores for the cardiac patients in this study? Provide a rationale for your answer . Answer: Parametric statistics, such as mean and SD , were conducted to describe CDS scores for study participants (see Table 1). CDS scores are interval-level data as indicated in Questions 5, so parametric statistics are appropriate for this level of data (Gray et al., 2017; Kim & Mallory, 2017). 7. Is the prevalence of depression linked to the NYHA class? Discuss the clinical importance of this result. Answer: The study narrative indicated that the prevalence of depression increased with the greater NYHA class. In NYHA class III, 64% of the subjects were depressed, whereas 11% of the subjects were depressed in NYHA class I. Thus, as the NYHA class increased, the number of sub- jects with depression increased. This is an expected finding because as the NYHA class increases, cardiac patients have more severe physical symptoms, which usually result in emotional distress, such as depression. Nurses need to actively assess cardiac patients for depression, especially those in higher NYHA classes, so they might be diagnosed and treated as needed. 8. What frequency and percent of cardiac patients in this study were not being treated with an antidepressant? Show your calculations and round your answer to the nearest whole percent (%). Answer: A total of 106 cardiac patients participated in this study. The sample included 15 patients who were receiving an antidepressant (see Table 1). The number of cardiac patients not treated for depression was 91 (106 – 15  91). The group percent is calculated by the following formula: (group frequency  total sample size)  100%. For this study, (91 patients  106 sample size)  100%  0.858  100%  85.8%  86%. The final answer is rounded to the nearest whole percent as directed in the question. You could have also subtracted the 14% of patients treated with antidepressants from 100% and obtained the 86% who were not treated with an antidepressant. 9. What was the purpose of the 6-minute walk test (6MWT)? Would the 6MWT be useful in clinical practice? Answer: Ha et al. (2018) stated, “The 6-min walk test (6MWT) is a measure of the submaximal, steady-state functional capacity” of cardiac patients. This test would be a quick, easy way to determine a cardiac patient’s functional status in a clinical setting. This functional status score could be used to determine the treatment plan to promote or maintain functional status of cardiac patients.

  4. Page 4
    Read page 4 as text

    AG 1-3 Answer Guidelines for Questions to Be Graded 10. How was exercise confidence measured in this study? What was the level of measurement for the exercise confidence variable in this study? Provide a rationale for your answer. Answer: Exercise confidence of the patients with heart failure (HF) in this study was measured with the Exercise Confidence Scale that included four subscales focused on walking, climbing, lifting objects of graded weight, and running (see the study narrative). This was a rating scale with values ranging from 0 to 100. The patients’ scores for the Total Exercise Confidence scale and the subscales were considered interval-level data and analyzed with parametric statistics, such as means and SDs (see the study narrative; Waltz et al., 2017 ).

  5. Page 5
    Read page 5 as text

    AG 2-1 EXERCISE 2 Identifying Probability and Nonprobability Sampling Methods in Studies Answer Guidelines for Questions to Be Graded Directions: Answer the following questions with a clear, appropriate response. For each question that includes an excerpt about sampling from a research article, provide the following information: (1) decide whether the sampling method presented is either a probability or nonprobability sampling method ; (2) identify the specific sampling method used—that is, convenience, quota, purposive, network, or theoretical sampling for nonprobability samples or simple random, stratified random, cluster, or systematic sampling for probability samples (see Table 2-1 ); and (3) provide a rationale for the sampling method you selected. Some of the examples might include more than one sampling method to obtain the study sample. Answers for Questions to be Graded Questions are in bold followed by answers. 1. Study excerpt: “Participants in this study were all women who had had, or might have, sex with men; that is, women not in exclusively same-sex partnerships. All participants were 18 to 35 years old, English-speaking, and attended a large, public, Midwestern university. More than 2,000 women completed Phase I of the study, a survey; of these approximately 900 volunteered to participate in the Phase II interviews. Women were selected from this pool to be interviewed using a purposive maximum variation sampling strategy . . . Based on survey data, women were sampled based on variation in their knowledge and use of emergency contraception and on some aspects of sexual history . . . Theoretical sampling was used to select potential participants based on the data as they were collected, including women who had experienced a pregnancy termination or ‘false alarm’ pregnancy . . . This process yielded a sample of 35 women who represented a wide range of experiences. When saturation was reached, recruitment and data collection ended.” The data for this study were obtained from a larger mixed methods study that included quantitative survey data and qualitative narrative interview data focused on collegiate women’s sexual knowledge and behavior. Source: Loew, N., Mackin, M. L., & Ayres, L. (2018). Collegiate women’s defi nitions of responsible sexual behavior. Western Journal of Nursing Research, 40 (8), 1148–1162. Excerpt from page 1151. Answer: Nonprobability convenience, purposive, and theoretical sampling methods. Loew et al. (2018) conducted a mixed method study with collegiate women to define responsible sexual be- havior. Phase I was the quantitative part of the study; collegiate women consented to participate in a survey, which indicates convenience sampling was used. In Phase II (the qualitative part of the study), women were purposely selected based on their variation of sexual knowledge and history. Theoretical sampling was used to ensure saturation of the qualitative data for defining responsible sexual behavior (Creswell & Clark, 2018; Gray, Grove, & Sutherland, 2017).

  6. Page 6
    Read page 6 as text

    AG 2-2 Answer Guidelines for Questions to Be Graded 2. Study excerpt: “This study was a part of a multi-site prospective observational study of COPD [chronic obstructive pulmonary disease] patients to explore the relationship be- tween depression, inflammation, and functional status . . . Participants were recruited from various sources including outpatient clinics from three medical centers, pulmonary reha- bilitation programs, a research database maintained by the investigators, queries of medi- cal records and pulmonary function tests, Better Breathers Club, community pulmonary practices, advertisements, study website, and other referrals.” A total of 282 patients with COPD were included in the study. Source: Lee, J., Nguyen, H. Q., Jarrett, M. E., Mitchell, P. H., Pike, K. C., & Fan, V. S. (2018). Effect of symptoms on physical performance in COPD. Heart & Lung, 47 (2), 149–156. Ex- cerpt from page 150. Answer: Nonprobability sample of convenience. Lee et al. (2018) recruited potential study participants from a variety of settings to obtain an adequate number of patients with COPD for this study. The potential participants were accessible, available, and willing to participate in the study, which is descriptive of a sample of convenience (Grove & Gray, 2019; Heavey, 2019; Kazdin, 2017). 3. Study excerpt: The focus of this quasi-experimental study was to determine the effect of a three-stage nursing intervention to increase women’s participation in Pap smear screen- ing. Using “random sampling methodology, each apartment in the target area was identi- fi ed by a number. Numbers were then drawn from a random numbers table. Women were contacted by home visits. . . . By the end of this stage, 237 participants had completed the pre-test.” Source: Guvenc, G., Akyuz, A., & Yenen, M. C. (2013). Effectiveness of nursing interven- tions to increase Pap smear test screening. Research in Nursing & Health, 36 (2), 146–157. Excerpt from page 148. Answer: Probability, simple random sampling method. The excerpt states that random sampling was used in the study. The sampling frame was the apartments in a targeted area, and simple random sampling with a random numbers table was used to select the apartments with their occupants for this study. A total of 237 women from the randomly selected apartments were included in the study. 4. Was the sample identified in the Guvenc et al. (2013) study in Question 3 representative of the population of women requiring a Pap smear test in a target area? Provide a rationale for your answer. Answer: Yes, the sample is probably representative of the study population in the selected target area. The simple random sampling process increases the representativeness of the sample and decreases the potential for sampling error or bias. The sample size of 237 women was adequate for this study ( Gray et al., 2017 ). 5. “Participants were recruited from January 2003 through November 2007 during their initial evaluation at the Pediatric Pain Management Clinic at Children’s Hospital Los Angeles . . . Ninety-six child-caregiver dyads were approached for study participant in order to obtain 65 sets of completed measures, resulting in a 68% participant rate. Nineteen sets were not returned, 7 sets did not have a complete child battery, 1 set did not have a complete caregiver battery, 2 child-caregiver dyads withdrew, and 2 families declined to participate . . . Children were considered eligible for the study if they were English speaking, between the ages of 8 and 18, had a diagnosis of chronic pain, and had a caregiver present.” Source: Yetwin, A. K., Mahrer, N. E., John, C., & Gold, J. I. (2018). Does pain intensity mat- ter? The relation between coping and quality of life in pediatric patients with chronic pain. Journal of Pediatric Nursing, 40 (3), 7–13. Excerpt from page 8.

  7. Page 7
    Read page 7 as text

    AG 2-3 Answer Guidelines for Questions to Be Graded Answer: Nonprobability, convenience sampling method. Participants were children initially at- tending a chronic pain management clinic. The child–caregiver dyads were approached and asked to participate until a sample of 65 dyads was obtained. This is consistent with convenience sam- pling, because the participants were available and willing to participate ( Grove & Gray, 2019 ). 6. Study excerpt: “Participants were 559 substance users recruited from multiple sources (parks, streets, prisons, methadone maintenance therapy, and drop in centers) . . . a nonprobability sampling technique that is appropriate to use in research when the members of a population are difficult to locate. In this research, we collected data from substance abusers . . . and then asked those individuals to locate other substance abusers whom they knew.” Source: Barati, M., Ahmadpanah, M., & Soltanian, A. R. (2014). Prevalence and factors associated with methamphetamine use among adult substance abusers. Journal of Research in Health Sciences, 14 (3), 221–226. Excerpt from page 222. Answer: Nonprobability network or snowball sampling method was identified by the Barati et al. (2014) in their study. However, the initial sampling process is probably one of convenience in which available participants are recruited from multiple settings, and then network or snowball sampling is used when participants are asked to identify additional substance abusers. Student answers might include both convenience and network or snowball sampling methods or just snowball sampling as indicated in the original article. 7. Study excerpt: Mansfield et al. (2018) conducted a correlational study to examine the association between parental knowledge of human papillomavirus (HPV) and their inten- tions to have their daughters vaccinated. “This study used HINTS [Health Information National Trends Survey] 2006-2007 because it was the only data set that assessed the outcome variable, intention to vaccinate for HPV . . . HINTS’s probability-based sample design used a random-digit dialing to conduct telephone surveys and a nationwide address list to administer surveys via mail. A subsampling screening tool, Westat’s Telephone Research Center (TRC), was used to identify working residential numbers. A total of 3,767 telephone interviews were then completed, and 325 were partially completed ( n  4,092); 3,473 mail surveys were completed and 109 were partially completed ( n  3,582). The fi nal total sample was 7,674 participants.” Source: Mansfield, L. N., Onsomu, E. O., Merwin, E., Hall, N. M., & Harper-Harrison, A. (2018). Association between parental HPV knowledge and intentions to have their daugh- ters vaccinated. Western Journal of Nursing Research, 40 (4), 481–501. Excerpt from page 481. Answer: Probability, simple random sampling method. Mansfield et al. (2018) clearly identified they used a probability, random sampling method in their study ( Grove & Gray, 2019 ). Potential participants were identified from the HINTS’s database, and random-digit dialing was used to conduct the telephone surveys. A national address list was used for mailing surveys to partici- pants, but it is unclear how the participants were selected from the address list or if all on the list were mailed surveys. 8. Was the sample identified in the Mansfield et al. (2018) study in Question 7 representative of parents’ intentions to have their daughters vaccinated? Provide a rationale for your answer. Answer: The sample is representative of the target population of parents’ intentions to have their daughters vaccinated. Simple random sampling provides strong representation of the target population that increases with sample size (see Table 2-1 ; Grove & Gray, 2019 ). The participants were randomly selected from a national database (HINTS) to conduct telephone surveys, and a national address list was used for mailing surveys, which documents the random sampling method. The total sample was extremely strong with 7674 participants. Recruiting participants

  8. Page 8
    Read page 8 as text

    AG 2-4 Answer Guidelines for Questions to Be Graded from the HINTS database is appropriate because it was the only data set that included the out- come variable of intention to vaccinate for HPV. This large simple random sample is representa- tive of the target population with limited potential for sampling error or bias ( Gray et al., 2017 ). 9. Study excerpt: Initially, participants were selected in a purposeful manner “based on their familiarity with, interest in, and willingness to reflect and discuss their hope experience.” Additional sampling was done to achieve “theoretical saturation that was defined as theo- retical completeness in which no new properties of the categories were identified.” Source: Bally, J. M., Duggleby, W., Holtslander, L., Mpofu, C., Spurr, S., Thomas, R., & Wright, K. (2014). Keeping hope possible: A grounded theory study of the hope experi- ence of parental caregivers who have children in treatment for cancer. Cancer Nursing, 37 (5), 363–372. Excerpt from page 364. Answer: Nonprobability, purposive and theoretical sampling methods. The initial sampling process is purposive, and Bally et al. (2014) clearly indicate the type of person they want to par- ticipate to provide the information-rich data needed in their study. Theoretical sampling is done to provide theoretical saturation and to ensure the credibility and quality of the theory devel- oped in this grounded theory study ( Creswell & Porth, 2018 ; Gray et al., 2017 ). 10. Study excerpt: Macartney and colleagues (2018) studied the concussion symptoms in 136 adolescents, 74 female and 62 male. “A retrospective chart review was completed between 11/21/2014 to 11/20/2015. A purposive sample of all patients who visited the CHEO [Children’s Hospital of Eastern Ontario] concussion clinic during the study period [was] included. Patients were excluded if symptoms records were not documented . . . CHEO’s concussion clinic opened in the fall of 2014. The clinic provides care to patients less than nineteen years old who remain symptomatic at least four weeks post injury.” Source: Macartney, G., Simoncic, V., Goulet, K., & Aglipay, M. (2018). Concussion symptoms prevalence, severity and trajectory: Implications for nursing practice. Journal of Pediatric Nursing, 40 (1), 58–62. Excerpt from page 59. Answer: Nonprobability, purposive sampling method. Macartney et al. (2018) stated that a pur- posive sampling method was used to select study participants who visited the CHEO concus- sion clinic during the study time period. Students might have identified the original sample of adolescents who visited the CHEO concussion clinic as comprising a sample of convenience because the adolescents were accessible and willing to participate in the study.

  9. Page 9
    Read page 9 as text

    AG 3-1 Answer Guidelines for Questions to Be Graded Understanding the Sampling Section of a Research Report: Population, Sampling Criteria, Sample Size, Refusal Rate, and Attrition Rate EXERCISE 3 Questions are in bold followed by answers. 1. Did the study include sampling inclusion criteria? What were those criteria? Were these criteria appropriate for this study? Answer: Yes, the mother – child dyads were eligible for this study if they met the following criteria: “(a) the targeted child was between the ages of 3 to 5 years at the time of data collection between March 2010 and March 2011, (b) the mother had primary custody or regular visitation with the child, (c) the dyad lived in state and/or was able to meet in state for the data collection, and (d) the mother participated in the Minding the Baby (MTB) home visiting program or the control condition beyond the initial consent period” (Ordway et al., 2014, p, 5). Yes, these criteria were appropriate and provided a fairly homogenous sample of mother – child dyads to determine the effectiveness of the MTB program ( Grove, Gray, & Sutherland, 2017 ). 2. Sampling inclusion and exclusion criteria are developed to determine what type of population? What is the accessible population? Document your answer. Answer: The sampling or eligibility criteria are developed to determine the target population for a study (see Figure 3-1 ). The accessible population is the portion of the target population to which the researcher has reasonable access ( Grove & Gray, 2019 ; Kim & Mallory, 2017 ; Knapp, 2017). (Note: Students might include a variety of sources for their documentation.) 3. How many of the mother – child dyads with children 3 to 5 years old were available for contact for the follow-up study? How many mother – child dyads were included in the initial intervention group for the follow-up study? What frequency and percent of the mother – child dyads were unable to be contacted in the intervention group? Answer: A total of 71 mother – child dyads with children 3 to 5 years old were available. The initial intervention group included 36 mother – child dyads. Figure 1 indicates 7 mother – child dyads could not be contacted. Percentage not contacted  (7  36)  100%  0.1944  100%  19.44%  19.4% or 19%. 4. Of the 36 mother – child dyads in the intervention group, how many were included in the follow-up study? How many mother – child dyads did not participate in the study? Was a rationale provided for their not participating? If so, what were the reasons and did they seem appropriate or unusual? Answer: The final intervention group included 24 mother – child dyads. Figure 1 indicates 12 dyads did not participate because 5 did not meet sample inclusion criteria, and 7 could not be contacted. These are common and appropriate reasons for not including participants in a

  10. Page 10
    Read page 10 as text

    AG 3-2 Answer Guidelines for Questions to Be Graded study. The final sample of 24 mother – child dyads in the intervention group was adequate because the power analysis determined that 17 dyads were needed for each group. 5. What was the attrition number and rate for this study? Provide a rationale for your answer. Answer: The study had no or zero (0%) attrition because all mother – child dyads included in the final intervention and control groups were included in the follow-up analyses as indicated in Figure 1. 6. Calculate the attrition rate for a study that included a sample size of 276 subjects and 10 subjects became too ill to continue in the study, 5 could not be contacted after hospital discharge, and 4 died. Round your answer to the nearest tenth of a percent. Answer: Attrition frequency  10  5  4  19. Attrition rate  (Attrition frequency  sample size)  100%  (19  276) x 100%  0.0688  100%  6.88%  6.9%. The small attrition rate is a study strength. 7. What was the total sample for this study? Answer: The total sample is 50 mother – child dyads as indicated in the study results section and can be calculated from Figure 1. Intervention group n  24 dyads and control group n  26 dyads total to a sample size of 50 mother – child dyads. 8. The intervention and control groups had unequal numbers. Is this a study strength or weakness? Provide a rationale for your answer. Answer: The comparison of intervention and control groups is stronger if these groups are equal size for study participants ( Holmes, 2018 ). The groups are approximately equal in size, so this is more a study strength than weakness and probably did not affect the study results. 9. Discuss the results of the power analysis conducted for this study. Was the sample size adequate for this study? Provide a rationale for your answers. Answer: The study sample size was calculated by a power analysis with α  0.05 and 80% power but the effect sized for the analysis was not identified. The power analysis indicated that 17 dyads were needed for each group, which indicates the sizes of the groups were adequate if the power analysis was correctly conducted with an appropriate effect size. Based on the dyads recruited, researchers noted there was potential for 29%, attrition. Because the study had no at- trition, the sample size would probably be considered adequate. However, the variable parental reflective functioning (RF) was not significantly different for the intervention and control groups, so the researchers need to investigate this result ( Gray et al., 2017 ; Holmes, 2018 ; Taylor & Spurlock, 2018 ). 10. Are the findings from this study ready for use in practice? Provide a rationale for your answer. Answer: Students’ answers might vary. The results of this study were mixed, and the sample size was small. The MTB intervention resulted in a significant decrease in externalizing child behavior problems but was not significant for the parental RF. Ordway et al. (2014) recom- mended additional research to determine the effect of the MTB intervention on parental RF. However, you might consider the MTB intervention that has benefits and minimal risks to be ready for use in practice to improve child behavior problems.

  11. Page 11
    Read page 11 as text

    Answer Guidelines for Questions to Be Graded AG 4-1 Understanding Reliability of Measurement Methods EXERCISE 4 Questions are in bold followed by answers. 1. Based on the information provided from the Williams et al. (2017) study, which of the two scales have the higher Cronbach’s alpha coefficient for this study? Which scale has the greater random error? Provide rationales for your answers. Answer: The Personal Experiences Checklist (PECK) has the higher internal consistency reliability for the overall scale of   0.94, because this value is closer to 1.00 (a perfect reliability value). The greater the Cronbach’s alpha coefficient is or the closer the value is to 1.00, the stronger the reliability of the scale and the lower the random error ( Waltz, Strickland, & Lenz, 2017 ). The Center for Epidemiologic Studies Depression Survey (CESD-10) has a reliability of   0.86 for the current study. Thus, the CESD-10 has the lower internal consistency reliability value and the higher random error. 2. Would you consider the CESD-10 a reliable measure of depression in this population of ninth grade adolescents? Provide a rationale for your answer. Answer: Yes, the CESD-10 was a reliable scale for measuring depression in this population of ninth grade adolescents because the scale’s Cronbach’s alpha coefficient was 0.86, which indicates strong reliability with limited random error ( Gray, Grove, & Sutherland, 2017 ). In addi- tion, the CESD-10 has been used in other studies of adolescents with a reliability   0.85 ( Brandley, Bagnell, & Brannen, 2010 ). 3. What types of reliability testing were provided for the cyberbullying subscale of the PECK? Were the reliability values acceptable? Provide a rationale for your answer. Answer: Cyberbullying subscale was examined for stability reliability with test-retest reliability ( r  0.86) and for internal consistency reliability with Cronbach’s alpha (   0.90). Both of these reliability values are strong because they are greater than 0.80 and are close to 1.00, a perfect reliability value (Bialocerkowski et al., 2010; Waltz et al., 2017). These values support the reliabil- ity of the cyberbullying subscale in this study. 4. Calculate the percentage of random error for the cyberbullying subscale of the PECK. Show your calculations and round to the nearest whole percent. Answer: Percentage of random error  (1.00 – 0.90 2 )  100%  (1.00 – 0.81)  100%  0.19  100%  19% 5. What are the Cronbach’s alphas for the four subscales for the PECK used to measure bullying in the Williams et al., (2017) study? Did these subscales have adequate reliability in this study? Provide a rationale for your answer. Answer: Three of the subscales of the PECK had Cronbach’s alphas  0.90 (verbal/relational bullying   0.90, physical bullying   0.91, and cyberbullying   0.90). The cultural bullying

  12. Page 12
    Read page 12 as text

    AG 4-2 Answer Guidelines for Questions to Be Graded subscale also had an adequate reliability (   0.78) in this study. Usually subscales have lower reliability values than the total scale, such as the PECK in this study, because they include fewer items. The cultural bullying subscale included only four items, which probably resulted in the lower Cronbach alpha value for this subscale. The PECK and the subscales in this study had strong reliability values, supporting the use of this scale in this population ( Gray et al., 2017 ; Waltz et al., 2017 ). 6. Which subscale of the PECK had the lowest Cronbach’s alpha? What is the random error for this subscale? Show your calculations and round to the nearest whole percent. Answer: The cultural bullying subscale had the lowest reliability (   0.78) of the PECK for this study. Percentage of random error  (1.00 0.78 2 )  100%  (1.00 0.6084)  100%  0.3916  100%  0.39  100%  39% 7. Williams et al. (2017, p. 25) reported the PECK “was previously used in adolescents’ ages 8 to 15.” Based on readability, do you think the PECK is acceptable for the population in this study? Provide a rationale for your answer. Answer: Yes, the PECK probably has an acceptable readability level for the participants in this study based on the strong homogeneity reliability (   0.94), and the scale has been used in a previous study with adolescents. However, the discussion of the PECK would have been strengthened if the researchers had provided a reading level for this scale ( Gray et al., 2017 ; Waltz et al., 2017 ). 8. Could you assume that the PECK would be a reliable scale to measure bullying in college students? Provide a rationale for your answer. Answer: No, you cannot assume that the PECK would be a reliable scale to measure bullying in college students. The reliability of a scale is specific to each study and should be calculated for the scale in this study ( Waltz et al., 2017 ). In addition, the studies previously conducted were with adolescents and not young adults, so the reliability of the scale might vary with the age of the study participants. 9. Did Williams et al. (2017) provide a clear discussion of the reliability of the CESD-10 and PECK scales in their study? Provide a rationale for your answer. Answer: Yes, these researchers provided a clear discussion of instrument reliability in their study. The internal consistency reliability for the CESD-10 and PECK was provided using the Cronbach’s alpha that is the most accepted form of homogeneity testing for multi-item scales with measurement at the interval level. Cronbach’s alphas and test-retest r values were also provided for the subscales of the PECK, which demonstrated strong reliability in measuring the sub-concepts of bullying. However, the reliability discussion would have been strengthened by including more information on different types of reliability testing from previous studies ( Gray et al., 2017 ). 10. Was the collection of saliva for cortisol values precise? Provide a rationale for your answer. Answer: Yes, the collection of saliva for cortisol values appeared precise. The researchers provided details about where, when, and how they collected the salivary specimens from the ninth graders. Students received the same instructions in preparation for the collection of the specimens and during the collection process. The researchers followed detailed steps to ensure the salivary specimens were consistently collected, stored, and transported for analysis ( Ryan-Wenger, 2017 ).

  13. Page 13
    Read page 13 as text

    Answer Guidelines for Questions to Be Graded AG 5-1 EXERCISE 5 Understanding Validity of Measurement Methods Questions are in bold followed by answers. 1. Discuss the importance of a measurement method’s validity in a study. Answer: The validity of a measurement method is essential to accurately measure a concept or variable in a study. A scale with reliability and validity makes it possible to obtain quality measures of a variable, and the measurements will yield accurate data that reflect what is being measured. Without quality measurement methods, the results and the findings from a study are questionable (Grove & Gray, 2019; Waltz, Strickland, & Lenz, 2017). 2. Does the CESD-10 scale have evidence of successive verification validity in the Williams et al. (2017) study? Provide a rationale for your answer. Answer: Yes, the CESD-10 has evidence of successive verification validity because the scale was used in a previous study of 156 adolescents conducted by Bradley, Bagnell, and Brannen (2010). Williams et al. (2017) also indicate that the CESD-10 was applicable to adolescent populations, and a score of 10 was clinical meaningful, which indicate previous research using this scale ( Waltz et al., 2017 ). 3. Did the CESD-10 have evidence of criterion-related validity in the Williams et al. (2017) study? Provide a rationale for your answer. Answer: Yes, the CESD-10 has evidence of criterion-related validity as indicated by the following statement: “A score of 10 or greater indicates need for referral (clinically meaningful)” ( Williams et al., 2017 , p. 25). This statement indicates that the CESD-10 can be used to predict depression in adolescents, and a score of 10 indicates a need for referral. 4. Discuss the evidence of validity obtained by factor analysis that is provided for the CESD- 10 in the Williams et al. (2017) study. Answer: Williams et al. (2017 , p. 25) reported that the CESD-10 had been used “to measure four factors [sub-concepts] related to depressive symptoms including: positive/negative affect, somatic symptoms, retarded activity, and interpersonal issues.” Factor analysis is conducted to statistically identify the factors or sub-concepts of a concept, which were identified for depres- sion in this study ( Waltz et al., 2017 ). 5. Provide an example of how convergent validity might have been examined for the CESD-10. Answer: Convergent validity is examined by administering two scales measuring the same concept to a group at the same time. The subjects’ scores on the scales should be positively correlated if the scales are measuring the same concept. For example, the CESD-10 and the

  14. Page 14
    Read page 14 as text

    AG 5-2 Answer Guidelines for Questions to Be Graded Patient Health Questionnaire (PHQ-9) are two scales, which measure depression in adolescents, that could be administer to a group of ninth graders in a school setting. If the scores from these two scales were positively correlated with values ranging from 0.40 to 0.80, the convergent validity of both scales would be strengthened ( Gray, Grove, & Sutherland, 2017 ; Waltz et al., 2017). Bhatta, Champion, Young, and Loika (2018) study focused on measuring depression in adolescents with the PHQ-9. This article is available to you in the online Article Library for this text. 6. What types of validity were discussed related to the CESD-10 scale in the Williams et al. (2017) study? Answer: Content validity, validity from factor analysis, successive verification validity, and predictive criterion-related validity were addressed for the CESD-10 scale. Content validity was evident in identifying the CESD-10 as measuring depressive symptoms. Content validity in- volves developing scale items to measure depression and determining the sub-concepts composed of selected scale items through literature and expert reviews. Factor analysis is conducted to statistically identify factors or sub-concepts of depression to determine if they match those identified through content validity ( Gray et al., 2017 ; Waltz et al., 2017 ). Successive verifi cation validity is expanded with additional studies conducted with the CESD-10. Criterion- related validity is evident with the score of 10 or greater, indicating a need for referral for depression. 7. Was the information on validity provided for the CESD-10 scale adequate in this study? Provide a rationale for your answer. Answer: The validity information for the CESD-10 was adequate but limited (Gray et al., 2017). The researchers needed to clearly identify and expand on the types of validity discussed in Question 6. Additional types of validity, such as convergent and divergent validity, would have strengthened the discussion of the CESD-10 in this study (see Table 5-1; Waltz et al., 2017). 8. Williams et al. (2017 , p. 25) reported, “Specimens were collected . . . under the principal investigator’s supervision. Specimens were kept on ice during data collection and then frozen at  20 °C until shipped to Salimetrics, LLC on dry ice.” What does this statement indicate about the precision and accuracy of the saliva specimens collected? Answer: The quote addressed the precision or consistency of the collection of saliva specimens that was supervised by one person, the principal investigator. The specimens were stored and then shipped for analysis in a precise manner. Precise data collection promotes the accuracy of the data collected. Keeping the specimens cold during data collection and frozen at a specifi c temperature promoted the accuracy of the specimens for analysis ( Gray et al., 2017 ; Ryan-Wenger, 2017 ). 9. How was accuracy determined in the measurement of cortisol levels in the Williams et al. (2017) study? Answer: The following quote clearly addresses the accuracy of the cortisol levels: “Duplicate analysis of saliva was completed with an ELISA/EIA assay, calibration range of 0.012–3.00 g/dL and a sensitivity of 0.003 g/dL to determine the cortisol levels. The expected range of salivary cortisol for adolescents aged 12–18 years is 0.021–0.883 g/dL for morning and 0.0–0.259 g/dL for afternoon, and the correlation between saliva and serum cortisol is r (47)  0.91, p 0.0001 (Salimetrics LLC, 2013)” (Williams et al., 2017, p. 25).

  15. Page 15
    Read page 15 as text

    AG 5-3 Answer Guidelines for Questions to Be Graded 10. Was the description of the measurement of cortisol diurnal rhythm adequate in this study? Provide a rationale for your answer. Answer: Williams et al. (2017) provided a quality description of the actions taken to ensure the cortisol levels and the cortisol diurnal rhythm (cortisol change from morning to afternoon) measurements were precise and accurate. The saliva specimens were collected, stored, and shipped in a precise manner to yield accurate results. Duplicate analysis of the saliva was conducted to determine the accuracy of the cortisol levels. The saliva and serum cortisol levels had a very strong correlation ( r  0.91) that was statistically significant at the p 0.0001, which supports the accuracy of the saliva cortisol level ( Ryan-Wenger, 2017 ).

  16. Page 16
    Read page 16 as text

    Answer Guidelines for Questions to Be Graded AG 6-1 Questions are in bold followed by answers. 1. What are the frequency and percentage of the COPD patients in the severe airflow limita- tion group who are employed in the Eckerblad et al. (2014) study? Answer: Seven (14%) of the COPD patients with severe airflow limitations were employed (see Table 1). 2. What percentage of the total sample is retired? What percentage of the total sample is on sick leave? Answer: In the total sample, 61% of the participants were retired, and 15% were on sick leave as identifi ed in the narrative of the study results. 3. What is the total sample size of this study? What frequency and percentage of the total sample were still employed? Show your calculations and round your answer to the nearest whole percent. Answer: The total sample size was 91 patients with COPD, as presented in the narrative of the study results. The number or frequency of subjects employed in the total sample is calculated by adding the frequencies from the moderate and the severe airflow limitation groups from Table 1. Calculation: Frequency employed  7 moderate group  7 severe group  14 employed COPD patients. The percentage of the sample employed  (Frequency employed  Sample size)  100%  (14  91)  100%  0.154  100%  15.4%  15% rounded to nearest whole percent. 4. What is the total percentage of the sample with a smoking history—either still smoking or former smokers? Is the smoking history for study participants clinically important? Provide a rationale for your answer. Answer: A total of 28% of the sample were still smoking  69% former smokers  97% smoking history (see narrative of results). The high smoking history (97%) for patients with COPD is very important clinically. This study adds to the research that links smoking with COPD. Because 28% of the study participants were still smoking, nurses need to be actively involved in assisting patients to reduce their number of cigarettes a day and ultimately quit smoking. Support needs to be provided to the former smokers to ensure they continue their nonsmoking behavior. 5. What are pack years of smoking? Is there a significant difference between the moderate and severe airflow limitation groups regarding pack years of smoking? Provide a rationale for your answer. Answer: Pack years of smoking  Mean (average) packs smoked per day  Number of years smoked (see footnote under Table 1 ). For example, a patient smoking a mean of 1.5 packs a day for 20 years will have a 30-pack-year smoking history. Pack years is a more accurate way to assess smoking history than just the number of years a person has smoked. No, the moderate and severe Understanding Frequencies and Percentages EXERCISE 6

72 more pages in the full document

Unlock it once and it stays in your library, ready to chat with or turn into flashcards.

Unlock the full document

More Nursing study materials

See all

Study this document with CramX

Once it is in your library, every tool can work from it.