Nurse Key

An exploratory descriptive survey design was used to obtain a convenience sample of women who self-report feelings of PPD across the past week [sample size n = 53]. Inclusion criteria were women between 2 weeks and 6 months postpartum who had been hospitalized for pregnancy complications. Women were excluded if they had a score of < 6 on the Edinburgh Postnatal Depression Scale (EPDS)…. EPDS is a widely used screening instrument to detect postpartum depression….

Maloni, Przeworski, and Damato (2013) studied women with postpartum depression (PPD) after pregnancy complications for the purpose of describing their barriers to treatment for PPD, use of online resources for assistance with PPD, and preference for Internet treatment for PPD. This study included a typical descriptive design; key aspects of this study’s design are presented in the following excerpt.

“Methods

In addition, a series of 26 descriptive questions assessed women’s barriers to PPD treatment, whether they sought information about depression after birth from any sources and their information seeking about PPD from the Internet, how often they sought the information, and whether the information was helpful. Questions were developed from review of the literature.… Content validity was established by a panel of four experts.… The survey was posted using a university-protected website using standardized software for surveys.”

(Maloni et al., 2013, pp. 91-92)

Maloni and associates (2013) clearly identified their study design as descriptive and indicated that the data were collected using an online survey. This type of design was appropriate to address the study purpose. The sample section was strengthened by using the EPDS to identify women with PPD and using the sample criteria to ensure that the women had been hospitalized for pregnancy complications. However, the sample size of 53 was small for a descriptive study. The 26-item questionnaire had content validity and was consistently implemented online using standard survey software. This typical descriptive design was implemented in a way to provide quality study findings.

Maloni and co-workers (2013) noted that of the 53 women who were surveyed because they reported PPD, 70% had major depression. The common barriers that prevented them from getting treatment included time and the stigma of PPD diagnosis. Over 90% of the women did use the Internet as a resource to learn about coping with PPD and expressed an interest in a web-based PPD treatment.

Comparative Descriptive Design

A comparative descriptive design is used to describe variables and examine differences in variables in two or more groups that occur naturally in a setting. A comparative descriptive design compares descriptive data obtained from different groups, which might have been formed using gender, age, educational level, medical diagnosis, or severity of illness. Figure 8-4 provides a diagram of this design’s structure.

Research Example

Comparative Descriptive Design

(Buet et al., 2013, pp. 72-73)

Buet and colleagues (2013) conducted a comparative descriptive study to describe and determine differences in the hand hygiene (HH) opportunities and adherence of clinical (e.g., nurses and physicians) and nonclinical (e.g., teachers and parents) caregivers for patients in pediatric extended-care facilities (ECFs). The following study excerpt includes key elements of this comparative descriptive design:

“Eight children across four pediatric ECFs were observed for a cumulative 128 hours, and all caregiver HH opportunities were characterized by the World Health Organization [WHO] ‘5 Moments for HH.’… A convenience sample of two children from each site (n = 8) was observed.… Four observers participated in two hours of didactic training and two hours of monitored practice observations at one of the four study sites to ensure consistent documentation and interpretation of observations. Observers learned how to accurately record HH opportunities and HH adherence using the WHO ‘5 Moments of HH’ data acquisition tool, discussed below. Throughout the study, regular debriefings were also held to review and discuss data recording.…The World Health Organization (WHO, 2009) ‘5 Moments for HH’ define points of contact when healthcare workers should perform HH: ‘before touching a patient, before clean/aseptic procedures, after body fluid exposure/risk, after touching a patient, and after touching patient surroundings…. During approximately 128 hours of observation, 865 HH opportunities were observed.”

Buet and associates (2013) clearly described the aspects of their study design but did not identify the specific type of design used in their study. The design was comparative descriptive because the HH opportunities and adherence for clinical and nonclinical caregivers were described and compared. The study included 128 hours of observation (16 hours per child) of 865 HH opportunities in four different ECF settings. Thus the sampling process was strong and seemed focused on accomplishing the study purpose. The data collectors were well trained and monitored to ensure consistent observation and recording of data. HH was measured using an observational tool based on international standards (WHO, 2009) for HH.

The purpose of a correlational design is to examine relationships between or among two or more variables in a single group in a study. This examination can occur at any of several levels— descriptive correlational, in which the researcher can seek to describe a relationship, predictive correlational, in which the researcher can predict relationships among variables, or the model testing design, in which all the relationships proposed by a theory are tested simultaneously.

Buet and co-workers (2013) found that the HH of the clinical caregivers was significantly higher than the nonclinical caregivers. However, the overall HH adherence for the clinical caregivers was only 43%. The low HH adherence suggested increased potential for transmission of infections among children in ECFs. Additional HH education is needed for clinical and nonclinical caregivers of these children to prevent future adverse events. Quality and Safety Education for Nurses (QSEN, 2013) implications from this study encourage nurses to follow evidence-based practice (EBP) guidelines in adhering to HH measures to ensure safe care of their patients and reduce their risk of potentially life-threatening infections (Sherwood & Barnsteiner, 2012).

Correlational Designs

In correlational designs, a large range in the variable scores is necessary to determine the existence of a relationship. Therefore the sample should reflect the full range of scores possible on the variables being measured. Some subjects should have very high scores and others very low scores, and the scores of the rest should be distributed throughout the possible range. Because of the need for a wide variation on scores, correlational studies generally require large sample sizes. Subjects are not divided into groups, because group differences are not examined. To determine the type of correlational design used in a published study, use the algorithm shown in Figure 8-5. More details on correlational designs referred to in this algorithm are available from other sources (Grove et al., 2013; Kerlinger & Lee, 2000).

Fig 8-5 Algorithm for determining the type of correlational design.

Descriptive Correlational Design

The purpose of a descriptive correlational design is to describe variables and examine relationships among these variables. Using this design facilitates the identification of many interrelationships in a situation (Figure 8-6). The study may examine variables in a situation that has already occurred or is currently occurring. Researchers make no attempt to control or manipulate the situation. As with descriptive studies, variables must be clearly identified and defined conceptually and operationally (see Chapter 5).

Research Example

Descriptive Correlational Design

This study used a descriptive, correlational design to examine the relationship between obesity and injury severity.… The convenience sample consisted of the 611 adolescent males, aged 11-17 years, who received medical attention for an injury at one of eight participating medical facilities. Exclusion criteria were adolescent males presenting with medical complaints unrelated to an injury (e.g., sore throat, dehydration, insect bite) and those who were classified as ‘special needs’ participants because of the disability affecting their mobility or requiring the use of an assistive device.… There were 20 medical facilities located throughout the 2010 National Boy Scout Jamboree. Each facility was equipped to manage both medical complaints and injuries.…”

Burns, Murrock, and Graor (2012) conducted a correlational study to examine the relationship between BMI and injury severity in adolescent males attending a National Boy Scout Jamboree. The key elements of this descriptive correlational design are presented in the following study excerpt.

“Design

(Burns et al., 2012, pp. 509–510)

“Measures

Past medical history, weight (in pounds) and height (in inches) were obtained from the HMR [health and medical record]. BMI [body mass index] and gender-specific BMI percentage were calculated electronically using online calculators from the Centers for Disease Control and Prevention and height and weight data. The BMI value was plotted on the CDC’s gender-specific BMI-for-age growth chart to obtain a percentile ranking (BMI-P)…. BMI-P defines four weight status categories: less than 5% is considered underweight, 5% to less than 85% is categorized healthy weight, 85% to less than 95% is the overweight category, and 95% or greater is categorized as obese. Age was measured in years and was self-reported.

Severity of injury was measured using the ESI [Emergency Severity Index] Version 4. This five-level triage rating scale was developed by the Agency for Healthcare Research and Quality and provides rapid, reproducible, clinically relevant stratification of patients into levels based on acuity and resource needs.… Training sessions were held for each medical facility to educate staff on the project, process, data collection techniques, and injury severity scoring methods.… All BMI and BMI-P values were recalculated to verify accuracy. To assess interrater reliability for injury severity scoring, ESI scores reported were compared with the primary researcher’s scores. When discrepancies were found, the primary researcher reviewed the treatment record to determine the most accurate score.”

(Burns et al., 2012, p. 510)

Descriptive Correlational Design

Burns and colleagues (2012) clearly identified their study design in their research report. The sampling method was a nonrandom sample of convenience that is commonly used in descriptive and correlational studies. Nonrandom sampling methods decrease the sample’s representativeness of the population; however, the sample size was large and included 20 medical facilities at a national event. The exclusion sampling criteria ensured that the subjects selected were most appropriate to address the study purpose. The adolescents’ height and weight were obtained from their medical records but the researchers did not indicate if these were reported or measured by the healthcare professionals. Self-reported height and weight for subjects could decrease the accuracy of the BMI and BMI-P calculated in a study. The BMI-P and severity injury scores were obtained using reliable and valid measurement methods, and the data from the medical facilities were checked for accuracy. The design of this study seemed strong and the knowledge generated provides a basis for future research.

Burns and associates (2012) found a significant relationship between BMI-P and injury severity. They noted that overweight/obese adolescents may have increased risks of serious injuries. Additional research is needed to examine the relationship of BMI to injury risk and to identify ways to prevent injuries in these adolescents. The findings from this study also emphasize the importance of healthy weight in adolescents to prevent health problems. QSEN (2013) implications are that evidence-based knowledge about the relationship between obesity and severity of injury provides nurses and students with information for educating adolescents to promote their health.

Predictive Correlational Design

The purpose of a predictive correlational design is to predict the value of one variable based on the values obtained for another variable or variables. Prediction is one approach to examining causal relationships between variables. Because causal phenomena are being examined, the terms dependent and independent are used to describe the variables. The variable to be predicted is classified as the dependent variable, and all other variables are independent or predictor variables. A predictive correlational design study attempts to predict the level of a dependent variable from the measured values of the independent variables. For example, the dependent variable of medication adherence could be predicted using the independent variables of age, number of medications, and medication knowledge of patients with congestive heart failure. The independent variables that are most effective in prediction are highly correlated with the dependent variable but are not highly correlated with other independent variables used in the study. The predictive correlational design structure is presented in Figure 8-7. Predictive correlational designs require the development of a theory-based mathematical hypothesis proposing variables expected to predict the dependent variable effectively. Researchers then use regression analysis to test the hypothesis (see Chapter 11).

Research Example

Predictive Correlational Design

“Design, Setting, and Sample

Coyle (2012) used a predictive correlational design to determine if depressive symptoms were predictive of self-care behaviors in adults who had suffered a myocardial infarction (MI). The following study excerpt presents key elements of this design.

A descriptive correlational design examined the relationship between the independent variable of depressive symptoms [agitation and loss of energy] and the dependent variable of self-care. Data were collected from 62 patients in one hospital, who were recovering from an MI in the metropolitan Washington, areaA….”

(Coyle, 2012, p. 128)

Measures

“Beck Depression Inventory II

Depressive symptoms were measured using the BDI-II [Beck Depression Inventory II], a well-validated, 21-item scale designed to measure self-reported depressive symptomatology.… Internal-consistency estimates coefficient alpha of the total scores were .92 for psychiatric outpatients and .93 for college students. Construct validity was .93 (p < .001) when correlated with the BDI-I. In this study, the BDI-II Cronbach’s alpha was .68 at baseline.”

(Coyle, 2012, p. 128)

“Health Behavior Scale

Self-care behaviors after an MI were measured by the Health Behavior Scale (HBS), developed specifically for measuring the extent to which persons with cardiac disease perform prescribed self-care behaviors.… This self-report, a 20-item instrument, assesses the degree to which patients perform five types of prescribed self-care (following diet, limiting smoking, performing activities, taking medications, and changing responses to stressful situations).… Cronbach’s alphas for different self-care behaviors ranged from .82 to .95. In this study, reliability was measured by Cronbach’s alpha and was .62 at 2 weeks and .71 at 30 days….Prior to hospital discharge, the Medical and Demographic Characteristics Questionnaire and BDI-II were administered by the researcher.… At 2 weeks and at 30 days after hospital discharge, participants were contacted by telephone to determine responses to the HBS.”

(Coyle, 2012, pp. 128-129)

Coyle (2012) might have identified her study design more clearly as predictive correlational but did clearly identify the dependent variable as self-care and the independent variables as depressive symptoms. The design also included the longitudinal measurement of self-care with the HBS at 2 weeks and 30 days. The design was appropriate to accomplish the study purpose. The sample of 62 subjects was adequate because the study findings indicated significant results. The BDI-II has documented reliability (Cronbach’s alphas > 0.7) and validity from previous studies, but the reliability of .68 was low in this study. Reliability indicates how consistently the scale measured depression and, in this study, it had 68% consistency and 32% error (1.00 − .68 = .32 × 100% = 32%; see Chapter 10). HBS had strong reliability in previous studies but the validity of the scale was not addressed. The reliability of HBS was limited at 2 weeks (62% reliable and 38% error) but acceptable at 30 days (71% reliable and 29% error). This study has a strong design with more strengths than weaknesses, and the findings are probably an accurate reflection of reality. The study needs to be replicated with stronger measurement methods and a larger sample.