Quantitative Research Example Project 801
math cartoon.jpg
Overview of experimental, quasi-experimental, descriptive
and observational research method designs using study examples.
By Group 2
Crystal Graham MSN
Deanna Hiott MSN
Donna Carrillo MSN
Kim Pickett MSN, FNP

Have you ever been confused about investigative research designs?

Do you just wish all that information was in one place!

Well, assistance is brought to you today by Group 2.

We have provided a table with an overview of experimental, quasi-experimental, descriptive and observational research method designs. The table also includes 4 studies that utilized the various research methods. All the investigations reviewed centered on cognitive behavior therapy and depression. A brief narrative follows the table with a few additional details. References are cited at the bottom of the page and classified by research method. Enjoy!

Design-------------------------------------------Features----------------------------------Strengths/Weaknesses of study design
Researcher states hypothesis
Selects people
Divides groups
Strengths: Randomizing increases validity

Weaknesses: More time consuming
More expensive, Sometimes randomizing not feasible
study presented
Cognitive Behavioral Therapy (CBT) vs. Interpersonal Therapy (IPT)
Randomized Controlled Trial
Randomization of participants
Manipulation of DV/IV
Study Results: Cognitive Behavioral Therapy and Interpersonal Therapy equal effectiveness

CBT superior in pts. with severe depression
CBT first line tx for severe depression
Used in social research
when randomization is
not feasible due to ethical, time or economic constraints
No Randomization
Some Manipulation
Some Control
No hypothesis
Resembles both,
quantitative & qualitative
Easy, timely, economical

No randomization,
Biologic scientists consider unreliable
Possibility of bias
study presented
Cognitive Behavioral Therapy (CBT) vs. Therapy as Usual (TAU)
Primary goal was to assess
training model

No Randomization
No manipulation
No Control
2 Predictions
No hypothesis

CBT taught
1 video of TAU assessed
2 videos of CBT assessed
1st at 6 months, 2nd at 1 year
Same patients, Same therapist
Ease of implementation - Treatment as usual (TAU)
patients became cognitive behavioral therapy (CBT)
patients with same therapist

Study Results: Outcomes showed CBT effective
Outcomes lined up with other RCT
No patient exclusions, just consent

No randomization, same patients, time may have improved not method
No controls or manipulation

Used for:
Trend analysis
Healthcare planning
Hypothesis generation
Resembles both
quantitative & qualitative
Surveillance programs
Generate hypothesis, identify risk factors, data available, inexpensive and efficient, depth of study

Difficult to organize, cause and effect unclear, casual inferences can be made, easily skewed
Study presented
Overview of issues
in delivery of
psychotherapeutic agents
Examination of literary evidence concerning CBT

Summary of evidence provided
Passive surveillance
Strengths: Data already available, indicates

Study Results: CBT very effective, Consider for many needs

Weaknesses: Cause and effect could be unclear however, in this study the literature review was large

Used to study/observe
No Control
No randomization
Lack of ID variable
Cohort studies
Case control studies
Ecological studies
Cross sectional studies
Case series studies
Case reports
Strengths: Helps study phenomena when ethically may cause harm, convenient, cost efficient

Weaknesses: Contributes less to empirical knowledge, lack of independent variable can lead to questions about causative factors
Study presented
Comparing face to face low intensity CBT with telephone low intensity CBT
2 or more CBT sessions
Phone CBT not inferior to FTF CBT Except for severe illness
Strengths: Large sample, Cost effective, Convenient, Better targeting, efficie increased access

Study Results: Phone CBT not inferior to F2F except for severe illness

Weaknesses: Can not be sure improvements were due to LICBT or natural improvements, Can not be sure of general representativeness, estimates of cost savings incomplete

Crystal Graham – Experimental Research Study Analysis

BACKGROUND - Interpersonal therapy (IPT) and cognitive-behavioral therapy (CBT) are effective short-term therapies for mild to moderate depression. However there is controversy regarding their effectiveness in severe depression. Luty et. al (2007) , performed a randomized controlled trial to compare the efficacy of interpersonal psychotherapy and CBT in people receiving out-patient treatment for depression and to explore response in severe depression.

Experimental Research ex. RCT

Has a hypothesis - In this study, authors hypothesized neither of the two psychotherapies would be particularly effective in participants with severe depression.

Manipulation - Use of either IPT or CBT in severely depressed outpatients. Does have independent and dependent variables

Control - After exclusion/inclusion criteria participants were randomized and of the 177 total 91 received IPT only and 86 received CBT only.

Randomization –Patients with a principal diagnosis of major depressive disorder were recruited from a wide variety of sources, including mental health out-patient clinics, general practitioners, self-referral and psychiatric emergency services. Patients were randomized to the two therapeutic interventions in a 1:1 ratio based on a computerized randomization sequence of permutated blocks of size 20.

Results – CBT and IPT comparable in effectiveness for outpatient short term therapy. CBT superior in patients with severe depression and should be first line treatment for severe depression.

Deanna Hiott - Quasi-experimental Research Design

Quasi-experimental research is a form of research often used in social science/psychology (Shuttleworth, 2008). Biological scientists consider this method unscientific and unreliable. However, it is useful for social measures and for studies where randomization is not feasible due to ethical, economical or time constraints (Newhouse, Dearholt, Poe, Pugh, & White, 2007). It resembles both quantitative and qualitative methods. Lastly, quasi-experimental methods also do not control for variables, this can make statistical analysis difficult (Shuttleworth, 2008).

The study that was evaluated was very interesting. The primary goal of this study was to assess the effectiveness of the training provided to community therapists on cognitive behavior therapy (CBT) (Simons et al., 2010). The depression levels of clients were measured with treatment as usual (TAU). Then the therapists in a community center were taught CBT. One TAU session was videotaped and then six months later and one year later, the same clients and therapists were videotaped again utilizing CBT. Clients also self evaluated their depression levels. At the conclusion of the investigation, the patients reported less depression and the therapists were satisfied with the CBT measures. Interestingly, these results do match previous randomly controlled trials assessing CBT (Simons et al., 2010).

Quasi-experimental design details:
Most common design: Nonequivalent Groups Design
Second most common: Regression-discontinuity Design
Other quasi-experimental designs: Proxy Pretest Design, Double Pretest Design, Nonequivalent Dependent Variables Design , Pattern Matching Design, Regression Point Displacement design
Good in social sciences where pre-selection and randomization is difficult
Often integrated with case studies
Allow statistical analysis
Less time and resources needed for experiment
No randomization
Do not examine pre-existing factors
No control group or control of variables
Can be a powerful tool, as long as shortcomings noted
Great for overview study then follow with case study or quantitative
(Trochim, 2006)

Donna Carrillo - Descriptive Research Design:
Descriptive research studies are observational studies designed to describe the patterns of disease occurrence in relation to variables such as person, place and time. They are most often the first step or initial inquiry into a new topic, event, disease, or condition (Grimes & Schulz, 2002). Scientists consider this to be an unreliable design. Three important uses of descriptive studies include trend analysis, healthcare planning and hypothesis generation. Good descriptive research should answer five basic “W” questions: who, what, when, where and finally, so what? It relies on quantitative and qualitative data.
The study I reviewed gives an overview of issues involved in the delivery of psycho-therapeutic interventions in specialty areas as well as in every day clinical practice. The goal of the study was to concentrate on the common health disorders: depression, suicidal ideations, anxiety, panic disorder, phobias, post-traumatic stress disorders, eating disorders, and several others. The intent was to assess the psychosocial intervention effectiveness. It effectively highlighted the extensive evidence for the effectiveness of cognitive behavior therapy. The implications were also noted that as health care professionals we need to consider delivering CBT and other evidence-based treatments that allow the greatest access to the most people for the most benefit.

Descriptive Research Design Details:
Types: Case Report, Case-series report, cross-sectional, surveillance, and ecological correlational studies.
Advantages: generate new hypothesis, identify risk factors, data already available and inexpensive and efficient to use, depth of study and opportunities.
Disadvantages: difficult to organize, causes and effect may be unclear, easy to draw casual inferences and easily skewed to fit the needs of the researcher.

Kim Pickett - Observational Research Designs
Observational research studies are a type of non-experimental research that is used to help gain knowledge pertaining to a particular phenomenon. Although experimental research is proven to have the most sound empirical foundation, observational research is often utilized to observe and study phenomenon that occur (Shuttleworth, 2009).

Types of observational research include cohort studies, case control studies, ecologic studies, cross-sectional studies, case series studies, and case reports (Gay, 2010). Due to the fact that observational studies often retrospectively review phenomenon that has already occurred, it does not compromise ethics, which is a strength of this type of research. As an example, Jepsen, Johnsen, Gillman and Sørensen (2004) cite the example of a study involving drug effects in pregnant women. It would be unethical encourage pregnant women to take drugs in pregnancy; therefore, an experimental design would not be appropriate in this case due to the obvious ethical concerns. However, this phenomenon can be studied retrospectively during observational research without causing egregious violations to ethics. Another benefit of this type of study design is the ability to form a hypothesis, which can further provide a starting point for an experimental research study (Gay, 2010).

Limitations of this study design relate to the fact that this contributes less to empirical research than experimental design (Gay, 2010). Furthermore, there is no control or randomization by the researcher, and lack of independent variable assignment makes it difficult to assume a cause and effect conclusion. The sample size may be small, such as is found in a case report of a solitary participant. Error or bias can also occur (Gay, 2010; Shuttleworth, 2009).

One observational study that was provided a clear illustration of observational research relates to patient- nurse staffing ratios and hospital readmission among pediatric patients (Tubbs-Cooley, Cimiotti, Silber, Sloane & Aiken, 2013). The aim of the study was to view the variables of hospital patient-to-nurse staffing ratios and readmission.

The authors studied the variables of hospital staffing ratio and hospital readmission within certain time frames among the pediatric population of four states using an (observational) cross-sectional study. To accomplish this, the researchers mailed surveys to random samples of RNs containing inquiries of their recollections of numbers of patients they had cared for on a particular day or week, and numbers of hours a week worked. The researchers also reviewed and compiled discharge data from state agencies for children ages 1-17 years, while excluding data that did not meet the criteria (such as missing admission or discharge dates). Overall, they had complete data for 90, 459 pediatric patients in four states. The researchers ultimately discovered that staffing ratio was inversely proportional to readmission within 15 to 30 days after discharge.

However, one major limitation of this study is that a causal relationship between nurse staffing and readmission's could not be determined, as there may have been other variables affecting the relationship. This is expected in this type of design, because the researchers cannot control variables. Another limitation to the study includes the self-reporting of number of patients by the nurses. The authors assert that self-report has been found to be reliable in the past (Tubbs-Cooley, Cimiotti, Silber, Sloane & Aiken, 2013), but it is another factor that is beyond the researchers' control.

An additional observational study, performed by Hammond et al. (2012) focuses on behavioral activation. The researchers (Hammond et al., 2012) sought to compare effectiveness of cognitive behavioral therapy (CBT) delivered either face to face (FTF) or over the telephone (OTT) in England. The sample size was 39,227 adults; the intervention was two or more sessions of CBT. Results of the study indicated that interventions with both FTF counseling and OTT counseling were effective, although adults with more severe symptoms as measured by the Patient Health Questionnaire, Generalized Anxiety Disorder questionnaire and Work and Social Adjustment Scales indicators noted improvement with FTF versus OTT counseling (Hammond, et al., 2012). Strengths included the knowledge that outcomes indicated CBT was effective in both modalities (Hammond, et al., 2012), which may result in more effective cost savings and access to health. One limitation was lack of randomization, control or manipulation as is inherent in this design protocol.

Experimental References:

Luty, S., Carter, J., McKenzie, J., Rae, A.M., Frampton, C., Mulder, R., Joyce, P. (2007). Randomized controlled trial of interpersonal psychotherapy and cognitive–behavioral therapy for depression. The British Journal of Psychiatry (190), 496-502.

Polit, D.F., & Beck, C.T. (2010). Nursing research appraising evidence for nursing practice. New York, NY. Lippincott Williams & Wilkins.

Quasi-Experimental References:

Newhouse, R. P., Dearholt, S. L., Poe, S. S., Pugh, L. C., & White, K. M. (2007). Johns Hopkins nursing evidence-based practice model and guide (1st ed.). Indianapolis, IN: Sigma Theta Tau International.

Shuttleworth, M. (2008, August). Quasi-experimental. Retrieved September 12, 2013, from http://explorable.com/quasi-experimental-design

Simons, A. D., Lewis, C. C., Murakami, J., Reid, M., Padesky, C. A., Montemarano, J.,...Beck, A. T. (2010). Training and dissemination of cognitive behavior therapy for depression in adults: A preliminaryexamination of therapist competence and client outcomes. Journal of Consulting and Clinical Psychology, 78(5), 751-756. doi:10.1037/a0020569

Trochim, W. M. (2006, October). Quasi-experimental design. Retrieved September 12, 2013, from http://www.socialresearchmethods.net/kb/quasiexp.php

Descriptive References:
Grimes, D.A. & Schulz, K.F. (2002) “Descriptive Studies: What they can and cannot do?” The Lancet, 359, 145-149.

Whitfield,G.& Williams, C. (2003). “The Evidence Base for Cognitive Behavioral Therapy in Depression: delivery in busy clinical settings.” Advances in Psychiatric Treatment (2003), vol. 9,21-30.

Observational References:
Gay, J. (2010, November). Clinical epidemiology and evidence based medicine glossary: Clinical study designs and methods terminology. Retrieved September 12, 2013, from http://www.vetmed.wsu.edu/courses-jmgay/GlossClinEpiEBM.htm

Hammond, G. C., Croudace, T. J., Muralikrishnan, R., Lafortune, L., Watson, A., McMillan-Shields, F., & Jones, P. B. (2012). Comparative effectiveness of cognitive therapies delivered face-to-face or over the telephone: An observational study using propensity methods. PLOS ONE, 7(9), 11-15. doi:10.1371/journal.pone.0042916

Jepsen, P., Johnsen, S., Gillman, M. & Sørensen, H. (2004). Interpretation of observational studies. Heart 2004; 90: 956-960. doi: 10.1136/hrt.2003.017269

Shuttleworth, M. (2009, February). Observational study. Retrieved September 13, 2013, from http://explorable.com/observational-study

Tubbs-Cooley, H., Cimiotti, J., Silber, J., Sloane, D., & Aiken, L. (2013). An observational study of nurse staffing ratios and hospital readmission among children readmitted for common conditions. BMJ Quality & Safety Online First: 7 May 2013.


Love, J. (2013, August). Snapshots cartoons. Retrieved from http://www.jasonlove.com/funny-cartoons/