Monday, January 20, 2014

Measures of Central Tendency and Dispersion with SPSS


Age

The sample consisted of 1500 participants, 1495 of which reported their ages between 18 and 89, an age range of 71 years. Five participants did not report their age. The average age of the participants was 46.22, and the median age was 45. Ages that fell one standard deviation above and below the mean were between approximately 29.5 to 62.9. The most frequently occurring age was 40 (n=48) at 3.2% of participants. The least frequent age was 80, with 2 participants (n = 2) or .3%.

Race

Of 1500 participants, all of which reported their race, 1181 (n = 1181) or 78.7% of the participants were White, 216 (n = 216), or 14.4% were Black, and 103 (n = 103), or 6.9% reported races other than White or Black. Although Table 1 lists statistics for both variables, because race is qualitative, the statistics are not meaningful for this variable. The most frequently occurring race (n = 1181) was White at 78.7%, The least frequent race was races identified as other than White or Black at 6.9% or 103 participants ( n = 103).

Tables and Figures

Table 1 presents the statistics associated with both variables, age and race, which includes mean, median, mode, standard deviation, range, minimum, and maximum statistics for age and race. Table 2 reports the frequencies and percentages associated with race categories, and Table 3 reports the frequencies and percentages of age categories. Figure 1 is a pie chart representing the races of the participants. It lists Whites as 78.73% of the total number of participants, Blacks as 14.40% and Other Races as 6.87%. Table 4 presents frequencies and percentages related to race as presented in the pie chart in Figure 1.

Syntax and output files for the variables race and age:



FREQUENCIES VARIABLES=RACE AGE

/STATISTICS=STDDEV RANGE MINIMUM MAXIMUM MEAN MEDIAN MODE SUM

/ORDER=ANALYSIS.


Frequencies

Notes
Output Created
15-JAN-2014 18:51:19
Comments

Input
Data
C:\Users\Deborah\Desktop\Stats\gss04student_corrrected.sav
Active Dataset
DataSet1
Filter
<none>
Weight
<none>
Split File
<none>
N of Rows in Working Data File
1500
Missing Value Handling
Definition of Missing
User-defined missing values are treated as missing.
Cases Used
Statistics are based on all cases with valid data.
Syntax
FREQUENCIES VARIABLES=RACE AGE
  /STATISTICS=STDDEV RANGE MINIMUM MAXIMUM MEAN MEDIAN MODE SUM
  /ORDER=ANALYSIS.
Resources
Processor Time
00:00:00.02
Elapsed Time
00:00:00.01



[DataSet1] C:\Users\Deborah\Desktop\Stats\gss04student_corrrected.sav


Table 1. 
Statistics

RACE OF RESPONDENT
AGE OF RESPONDENT
N
Valid
1500
1495
Missing
0
5
Mean
1.28
46.22
Median
1.00
45.00
Mode
1
40
Std. Deviation
.583
16.679
Range
2
71
Minimum
1
18
Maximum
3
89
Sum
1922
69105

Frequency Tables
Table 2.

RACE OF RESPONDENT

Frequency
Percent
Valid Percent
Cumulative Percent
Valid
WHITE
1181
78.7
78.7
78.7
BLACK
216
14.4
14.4
93.1
OTHER
103
6.9
6.9
100.0
Total
1500
100.0
100.0


Table 3.
AGE OF RESPONDENT

Frequency
Percent
Valid Percent
Cumulative Percent
Valid
18
12
.8
.8
.8
19
19
1.3
1.3
2.1
20
16
1.1
1.1
3.1
21
14
.9
.9
4.1
22
15
1.0
1.0
5.1
23
28
1.9
1.9
7.0
24
26
1.7
1.7
8.7
25
35
2.3
2.3
11.0
26
29
1.9
1.9
13.0
27
27
1.8
1.8
14.8
28
40
2.7
2.7
17.5
29
21
1.4
1.4
18.9
30
28
1.9
1.9
20.7
31
25
1.7
1.7
22.4
32
31
2.1
2.1
24.5
33
29
1.9
1.9
26.4
34
35
2.3
2.3
28.8
35
28
1.9
1.9
30.6
36
37
2.5
2.5
33.1
37
18
1.2
1.2
34.3
38
30
2.0
2.0
36.3
39
23
1.5
1.5
37.9
40
48
3.2
3.2
41.1
41
27
1.8
1.8
42.9
42
23
1.5
1.5
44.4
43
34
2.3
2.3
46.7
44
41
2.7
2.7
49.4
45
31
2.1
2.1
51.5
46
32
2.1
2.1
53.6
47
36
2.4
2.4
56.1
48
21
1.4
1.4
57.5
49
29
1.9
1.9
59.4
50
37
2.5
2.5
61.9
51
32
2.1
2.1
64.0
52
32
2.1
2.1
66.2
53
20
1.3
1.3
67.5
54
23
1.5
1.5
69.0
55
29
1.9
1.9
71.0
56
34
2.3
2.3
73.2
57
28
1.9
1.9
75.1
58
14
.9
.9
76.1
59
25
1.7
1.7
77.7
60
19
1.3
1.3
79.0
61
23
1.5
1.5
80.5
62
25
1.7
1.7
82.2
63
20
1.3
1.3
83.5
64
16
1.1
1.1
84.6
65
22
1.5
1.5
86.1
66
13
.9
.9
87.0
67
22
1.5
1.5
88.4
68
10
.7
.7
89.1
69
8
.5
.5
89.6
70
16
1.1
1.1
90.7
71
10
.7
.7
91.4
72
9
.6
.6
92.0
73
8
.5
.5
92.5
74
12
.8
.8
93.3
75
13
.9
.9
94.2
76
8
.5
.5
94.7
77
6
.4
.4
95.1
78
10
.7
.7
95.8
79
11
.7
.7
96.5
80
4
.3
.3
96.8
81
7
.5
.5
97.3
82
10
.7
.7
97.9
83
7
.5
.5
98.4
84
4
.3
.3
98.7
85
4
.3
.3
98.9
86
4
.3
.3
99.2
87
3
.2
.2
99.4
88
2
.1
.1
99.5
89
7
.5
.5
100.0
Total
1495
99.7
100.0

Missing
NA
5
.3


Total
1500
100.0



Syntax and output files for the pie chart for race:

FREQUENCIES VARIABLES=RACE
  /PIECHART PERCENT
  /ORDER=ANALYSIS.

Frequencies
Notes
Output Created
19-JAN-2014 13:50:29
Comments

Input
Data
C:\Users\Deborah\Desktop\Stats\gss04student_corrrected.sav
Active Dataset
DataSet1
Filter
<none>
Weight
<none>
Split File
<none>
N of Rows in Working Data File
1500
Missing Value Handling
Definition of Missing
User-defined missing values are treated as missing.
Cases Used
Statistics are based on all cases with valid data.
Syntax
FREQUENCIES VARIABLES=RACE
  /PIECHART PERCENT
  /ORDER=ANALYSIS.
Resources
Processor Time
00:00:00.20
Elapsed Time
00:00:00.38



[DataSet1] C:\Users\Deborah\Desktop\Stats\gss04student_corrrected.sav


Statistics
RACE OF RESPONDENT 
N
Valid
1500
Missing
0


Table 4.
RACE OF RESPONDENT

Frequency
Percent
Valid Percent
Cumulative Percent
Valid
WHITE
1181
78.7
78.7
78.7
BLACK
216
14.4
14.4
93.1
OTHER
103
6.9
6.9
100.0
Total
1500
100.0
100.0



Figure 1.




Posted by at No comments:

Replicating Milgram (Sampling)


The following post will critique the sampling strategy and sample size used in Burger (2009). An appropriate sampling strategy supports a study's generalizeability and part of the value of a study is its ability to make inferences about the population from which the sample is taken (Frankfort-Nachmias & Nachmias, 2008). A sample is representative of the population of focus and is used to make studying a larger population feasible.

Rather than focusing on creating a sample that was accurately representative of the general public, Burger's (2009) focus was to recreate an older study, and because of ethical constraints, was forced to utilize a careful and extensive process of participant selection. Miller (2009) suggested Burger's efforts may have been overdone, lacked effectiveness and cost the experiment too much in terms of findings and generalizeability.

Burger (2009) sought to replicate real life in an experimental setting, but his sampling strategy weakened the overall findings and limited their generalizeability to wider populations. In this laboratory experiment, participants cannot be considered representative of the population about which the research intends to generalize (Frankfort-Nachmias & Nachmias, 2008). The Burger study did not randomly select its participants; they were selected according to several criteria.

It is difficult to determine the effect of the selection and screening process in Burger (2009). A relatively large percentage of individuals were screened out because of psychological or emotional problems and other issues determined by the clinical psychologists who did the final screening. In all, out of the self-selected group of 180 (approximately) volunteers for participation, 30% were excluded. Then, out of the 123 who remained, 70 were chosen to participate, which ends up to be approximately 38% strategically selected from the initial volunteers for participation. Burger's (2009) sample could not be considered representative of the general population because of this highly selective process. Understandably, Burger did not use a random sampling strategy because of ethical concerns about the effects of exposure to the treatment. However, Miller (2009) suggested that perhaps Berger's ethical concerns were overemphasized and by screening out more than half of the volunteers, the cost to the study's precision was immense. After the extensive screening process, however, Berger used a random process to assign participants to one of two treatment groups.

It would appear that the sampling frame for Burger's study was not well-defined, and could not have been due to the constraints imposed by ethics of the psychological field. Although Burger identified his population as people in general, he could not select a sample that represented the general population. And although Burger defined his population, at least in theory, he could not simply randomly select his participants. His primary goal was to replicate Milgram.

Burger (2009) justified his sample size, claiming that although he failed to find significant difference between his and Milgram's study, it was not on account of a lack of power. He believed his sample size was within reasonable limits to the Cohen's recommendations for medium effect. Although the findings of Burger's study are legitimately comparable to Milgram's, his findings are not generalizeable to the general population.

Burger, J. M. (2009). Replicating Milgram: Would people still obey today? American Psychologist, 64(1), 1-11. doi:10.1037/a0010932

Frankfort-Nachmias, C. and Nachmias, D. (2008). Research methods in the social sciences (7th ed.). New York, NY: Worth Publishers.

Miller, A. G. (2009). Reflections on 'Replicating Milgram' (Burger, 2009). American Psychologist, 64(1), 20-27. doi:10.1037/a0014407
Posted by at No comments:

Measurement and Instruments for a Quantitative Research Plan




Measuring variables can be a complicated task, and selecting valid measures is equally complex (Sechrest, 2005). Determining the appropriate levels of measurement for variables ensures they will be correctly operationalized and measured (Frankfort-Nachmias & Nachmias, 2008). In addition, selecting a reliable assessment instrument ensures researchers' data is representative of the variables selected for the study (Frankfort-Nachmias & Nachmias, 2008). This correct representation affects the internal validity of the study (Sechrest, 2005). The purpose of this paper is to describe the levels of measurement important for my study and to discuss why they are important. It will explain how I will ensure reliability, content, empirical, and construct validity for the measurement utilized. It will discuss the strengths and limitations of the measurement instrument selected in terms of reliability and validity. Further, it will discuss the reliability and validity of the BDI-FastScreen, and the population for which this assessment instrument was designed.

Important Levels of Measurement

My study seeks to measure depression in patients with rare cancer types, so ordinal and interval levels of measurement are important. Variables that can be assessed incrementally are often measured at the ordinal level. They can be rank-ordered from highest levels to lowest levels. Unlike nominal levels of measurement, ordinal data propose that the variables can be presented in an order (Trochim, 2006). The psychological construct of depression can be ranked in terms of relative amounts (Brockopp, & Hastings-Tolsma, 2003). For example, depression can be ordered as mild, moderate, or severe. However, by operationalizing depression based on the scores of the Beck Depression Inventory Fast Screen for Medical Patients (BDI-FastScreen), the ordinal levels of measurement, mild, moderate, and severe, become interval level measurements.

Whereas the ordinal level of measurement characterizes continuous variables or increments that can be ordered from high to low, the interval level of measurement allows for a degree of difference between the increments. The data related to depression can be ordered according to the intensity of individuals' experiences (Brockopp, & Hastings-Tolsma, 2003). At ordinal levels, however, depression data cannot be associated with meaningful differences and distances between the categories of mild, moderate, and severe. No consistent numbers can be associated with the distance between the categories, so measuring depression cannot fulfill the criteria for interval levels of measurement unless it is operationalized based on scores from the BDI-FastScreen (Brockopp, & Hastings-Tolsma, 2003). The opererationalization of the ordinal levels of data according to the BDI-FastScreen allows me to capture more complex data on central tendency that are not available to ordinal levels of measurement. For example, it is possible to calculate the median and mode in a rank-ordered data, however, the interval level of measurement allows the calculation of a mean, as well as a range and standard deviation. Because of the additional information provided by the interval level of measurement gained by operationalizing levels of depression based onBDI-FastScreen scores, a more complex statistical analysis will be possible. This will enable my research study to provide more complex information on the variables (Frankfort-Nachmias & Nachmias, 2008).

Ensuring Content, Empirical, and Construct Validity

Content Validity
Consulting the literature is an important component to choosing and assessing an instrument that will actually measure what the researcher intends to measure (Brockopp & Hastings-Tolsma, 2003). Other researchers can provide examples of the successful use of instruments that have been of value in similar settings. A researcher's assessment of content validity will rely on prior research as well as the expert opinions of other scientists. If an instrument has not been used to measure the construct in a similar setting, it will be difficult to determine its content validity (Brockopp & Hastings-Tolsma, 2003).

I will be using the BDI-FastScreen, which has been utilized by healthcare professionals in medical and clinical settings, to assess behavioral and somatic symptoms associated with medical issues (Scheinthal, Steer, Giffin, & Beck, 2001). It is a reliable and valid information source when used to determine and differentiate depressive symptoms that result from medical conditions from depressive symptoms from other sources (Hennessey & Pallone, 2003; Whiston & Eder, 2003). It has been normed on populations similar to the population of focus in my study.

Empirical Validity

Empirical or predictive validity is the extent to which scores on one assessment correspond to the same behaviors measured with other assessment instruments. For an assessment to be empirically valid, statistical evidence must suggest the instrument measures what it is meant to measure (Trochim, 2006). The BDI-FastScreen was correlated with two other assessment instruments thatmeasure symptoms of depression and with the diagnostic criteria for depression in the Diagnostic and Statistical Manual of Mental Disorders IV-TR (DSM-IV-TR). The correlations were r = .62 with the Hospital Anxiety and Depression Scale and r = .86, when correlated with the Beck Anxiety Inventory for Primary Care. Correlation with the DSM-IV-TR was r = .69 (Hennessey & Pallone, 2003). Although the correlation to the less recent DSM may not continue to be relevant, the empirical validity of this instrument is based on these correlations (Hennessey & Pallone, 2003; Whiston & Eder, 2003).

Construct Validity

Construct validity is a primary concern for researchers (Schotte, Maes, Cluydts, De Doncker, & Cosyns, 1997). Assessing the validity of how well an instrument measures what it is supposed to measure is a critical component to the success of the researcher and the research (Cronbach, & Meehl, 1955). If the assessment measures something other than the construct of focus, the results may be misleading or meaningless. When assessing construct validity, it is critical to ensure that the instrument measuring the variable is actually measuring the variable intended for measurement and not some other construct (Cronbach, & Meehl, 1955). For example, if the goal is to measure depression, it would not be prudent or beneficial to measure fear.

Since the BDI-FastScreen was correlated with other assessments designed to measure depressive symptoms, I am confident with the construct validity demonstrated by this assessment. The BDI-FastScreen has been used to measure depression in a variety of medical conditions (Whiston & Eder, 2003) such as in patients with multiple sclerosis (Benedict, Fishman, McClellan, Bakshi, & Weinstock-Guttman, 2003), geriatric patients (Scheinthal, Steer, Giffin, & Beck, 2001), patients with chronic pain (Poole, Bramwell, & Murphy (2009) and cancer patients (Alacacıoğlu, Öztop, & Yılmaz, 2012).

Ensuring Reliability

Reliability is an estimation of the extent to which a measurement instrument will yield the same results upon reassessment (Trochim, 2006). The reliability of an instrument contributes to the level of usability for empirical research (Whiston, 2009). Further, it refers to the replicability andstability of a measurement and whether it will result in the same assessment in the same individuals when repeated (Frankfort-Nachmias & Nachmias, 2008). When determining the reliability of an assessment, a reliability coefficient of at least .80 indicates a trustworthy level of reliability (Trochim, 2006). In effect, the reliability of an instrument estimates the extent to which variance in response is real variance rather than an error in the implementation of the instrument.

If an assessment is reliable, its results are stable and relatively true (Whiston, 2009). Utilizing the BDI-FastScreen will provide a true and stable assessment of depression in the participants. The authors of the BDI-FastScreen based its reliability on four patient groups used as normative samples (Whiston & Eder, 2003). The test manual for the BDI-FastScreen supplied limited information on the assessment's reliability, however, the coefficient alphas for the four groups were .86; .85; .88; and .86 (Hennessey & Pallone, 2003). To gauge the assessment's test-retest reliability in my study, participants will be assessed with the BDI-FastScreen upon becoming participants in the study, after six weeks, and after twelve weeks of support group participation. This will ensure the test is a reliable indication of the participants' levels of depression.

Strengths and Limitations of the BDI-FastScreen

Strengths
The BDI-FastScreen is a user-friendly assessment for screening depressive symptoms in adolescents and adults with medical conditions. One critical benefit is that the BDI-FastScreen determines these symptoms as they relate to medical issues (Hennessey & Pallone, 2003; Whiston & Eder, 2003). In addition, this instrument is a pencil and paper self-report that is easily scored, and is a reliable resource for evaluating depressive symptoms in medical patients (Whiston & Eder, 2003). It has been shown to be an adequate measure in a variety of patient groups and across a wide range of ages and both sexes (Whiston & Eder, 2003). The BDI-FastScreen is one of the few assessments that is capable of identifying depressive symptoms in patients and differentiates between normal psychological reactions to cancer diagnoses and major depressive disorders (Whiston & Eder, 2003). The BDI-FastScreen is easy to administer and has the potential to assist clinicians in their overall client conceptualization. In addition, it eliminates clinician bias and the overestimation of the patient's improvement (Trivedi et al., 2004; Kramer, Owen, Wilson, & Thrush, 2003).

Limitations

Whiston and Eder (2003) suggested the high correlation between the Beck Anxiety Inventory for Primary Care and the BDI-FastScreen may mean that both instruments evaluate anxiety rather than depression. If this were the case, measuring depression accurately would not be accomplished by the use of this assessment. It may be that additional information and clarity is necessary to determine whether the BDI-FastScreen measures depression or anxiety. In addition, there has been some discussion as to whether the samples utilized limited the generalizeability of the instrument, since all participants were from the greater Philadelphia area, and the 268 participants randomly chosen were from only four medical settings.

Until additional studies are completed with representative samples this assessment instrument may not be representative of wider populations. In addition, no test-retest analysis was reported for the BDI-FastScreen and that deficiency may cause researchers to question this assessment's reliability (Hennessey & Pallone, 2003). For my study, it is essential that the depressive symptoms identified in the data collection are related to a medical condition rather than to other sources. The BDI- FastScreen was chosen for the purpose of identifying depressive symptoms as they relate to medical conditions. If the assessment does not assess levels of depression related to a cancer diagnosis, my research may be misleading or meaningless. The BDI-FastScreen may be subject to reporting bias from test takers minimizing responses or over reporting the severity of symptoms, which would reduce the test's validity (Trevedi et al., 2004).

An Appropriate Scale

The BDI-FastScreen is a seven item criterion-referenced instrument. Scores range from 0 to 21, with higher scores indicating an increased severity of depressive symptoms. The clinical interpretation of scores is attained by criterion-referenced methods and suggests scores between 4 and 8 are indicative of a mild major depressive disorders (MDD); scores ranging from 9-12 indicate moderate MDD; and scores of 13-21 suggest severe MDD (Hennessey & Pallone, 2003). The BDI-FastScreen incorporates a Likert scale to capture information on the responders' level of depressive symptoms. Likert scaling is a unidimensional scaling method (Trochim, 2006). This scale commonly utilizes questionnaires to collect information by allowing responders to provide answers that vary in intensity (Carifio & Perla, 2007). A Likert scale captures a range of intensity that represents responders' experiences and feelings since they can specify their level of agreement to the question or statement (Frankfort-Nachmias & Nachmias, 2008). Because the Likert scale contains a greater number of measurements within the assessment, these scales are more likely to be more reliable than single item assessments (Jamieson, 2004). This scale measures attitudes on the ordinal and interval levels of measurement (Frankfort-Nachmias & Nachmias, 2008). This scale is appropriate for my study since one of the primary goals is to obtain measurements of depressive symptoms as reported according to participant experiences and attitudes.

Scales allow researchers to define and categorize variables. According to Frankfort-Nachmias and Nachmias (2008), scales have properties which determine their appropriateness for use. Specific scales are utilized under certain conditions and with certain tests (Frankfort-Nachmias & Nachmias, 2008). Likert scaling assumes that distances on each item are equal, so ordinal and interval data are well-measured by this scale. Further, the Likert scale is designed to measure attitudes, which seems appropriate for my research study. Using a scale that is not appropriate for a test can result in inaccuracies and useless data (Frankfort-Nachmias & Nachmias, 2008). The Likert scale allows test-takers to respond to questions according to their personal experience. In addition, the results of a Likert scale, which is utilized in the BDI-FastScreen assists researchers in collecting ordinal data and transforming it into interval level data.

In my study, participants' levels of depression will be assessed with the BDI-FastScreen upon becoming participants, and at 6 weeks, and at 12 weeks of participation in either the face-to-face support group or the online, disease-specific support forum. My goal is to determine whether patients with rare cancers have lower depressive symptoms with participation in online, disease-specific forums than they do with participation in face-to-face support groups. The BDI-FastScreen will measure depressive symptoms three times during participation in the study.

Population Used

The population used for the BDI-FastScreen were medical patients from a variety of clinical settings. The BDI-FastScreen was normed on four groups (Whiston & Eder, 2003). The first was a group of 50 patients who had been referred to psychiatrists for consultation after hospitalization for a medical condition. The second group consisted of 94 patients referred from a family practice setting, and the third group was made up of 100 pediatric patients that ranged in age from 12 to 17 who were scheduled for routine medical appointments, and the fourth group consisted of 120 patients from a university clinic. This instrument was specifically designed for screening depressive symptoms related to medical circumstances (Hennessey & Pallone, 2003; Whiston & Eder, 2003). The BDI-FastScreen provides information for a unique population in which symptoms of depression from medical conditions must be differentiated from depressive symptoms from other sources. The population in this study is cancer patients between the ages of 40 and 60 who have been diagnosed with a rare cancer at least six months prior to becoming a participant in this study, and have participated in a face-to-face support group, or an online disease-specific online support forum for at least three months. A rare cancer type is defined as a cancer that makes up less than 1% of all cancer diagnoses.

Conclusion

Although measuring variables can be a complicated task, selecting a valid and reliable assessment instrument is foundational to a valid research design (Sechrest, 2005).  Determining appropriate levels of measurement is essential as is ensuring content, empirical, and construct validity as well as reliability.  These characteristics contribute to the usability and the practical implementation of the research.  Strengths and limitations exist for most assessment instruments, and it is important to identify how they might affect the study's results.  For my study, the BDI-FastScreen is a reliable and valid assessment instrument for measuring depressive symptoms in cancer patients.  It utilizes a Likert scale, which adequately measures attitudes on ordinal and interval levels of measurement (Frankfort-Nachmias & Nachmias, 2008).  The BDI-FastScreen was developed as a screening tool for depressive symptoms in the unique population of medical patients.

References

Alacacıoğlu, A., Öztop, İ., & Yılmaz, U. (2012). The Effect of Anxiety and Depression on Quality of Life in Turkish Non Small Lung Cancer Patients. Turk Toraks Dergisi / Turkish Thoracic Journal, 13(2), 50-55. doi:10.5152/ttd.2012.12

Benedict, R., Fishman, I., McClellan, M., Bakshi, R., & Weinstock-Guttman, B. (2003). Validity of the Beck Depression Inventory-Fast Screen in multiple sclerosis. Multiple Sclerosis, 9(4), 393-396. doi: 10.1191/1352458503ms902oa

Brockopp, D. Y., & Hastings-Tolsma, M. T. (2003). Chapter 6, Answering the research question: Quantitative designs. In Fundamentals of Nursing Research (pp. 193-220). Boston: Jones & Bartlett.

Carifio, J., & Perla, R. J. (2007). Ten Common Misunderstandings, Misconceptions, Persistent Myths and Urban Legends about Likert Scales and Likert Response Formats and their Antidotes. Journal of Social Sciences, 3(3), 106-116. doi: 10.3844/jssp.2007.106.116

Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52(4), 281-302. doi: 10.1037/h0040957

Frankfort-Nachmias, C., & Nachmias, D. (2008). Research Methods in the Social Sciences (7th ed.). New York: Worth.

Hennessy, J., J. & Pallone, N., J. (2003). Review of the BDI-FastScreen for Medical Patients. In B.S. Plake, J.C. Impara, & R.A. Spies (Eds.), The seventeenth mental measurements yearbook. Lincoln, NE: Buros Institute of Mental Measurements. Retrieved from http://web.ebscohost.com.ezp.waldenulibrary.org/

Jamieson, S. (2004). Likert scales: How to (ab)use them. Medical Education, 38(12), 1217-1218. doi: 10.1111/j.1365-2929.2004.02012.x

Kramer, T., Owen, R., Wilson, C., & Thrush, C. (2003). Relationship between self-report and clinician-rated impairment in depressed outpatients. Community Mental Health Journal, 39(4), 299-307.

Poole, H., Bramwell, R., & Murphy, P. (2009). The utility of the Beck Depression Inventory Fast Screen (BDI-FS) in a pain clinic population. European Journal of Pain, 13(8), 865-869. doi: 10.1016/j.ejpain.2008.09.017

Scheinthal, S. M., Steer, R. A., Giffin, L., & Beck, A. T. (2001). Evaluating geriatric medical outpatients with the Beck Depression Inventory-FastScreen for Medical Patients. Aging & Mental Health, 5(2), 143-148. doi: 10.1080/13607860120038320

Schotte, C. K., Maes, M., Cluydts, R., De Doncker, D., & Cosyns, P. (1997). Construct validity of the Beck Depression Inventory in a depressive population. Journal of Affective Disorders, 46(2), 115-125.

Sechrest, L. (2005). Validity of Measures Is No Simple Matter. Health Services Research, 40(5p2), 1584-1604. doi: 10.1111/j.1475-6773.2005.00443.x

Trivedi, M. H., Rush, A. J., Ibrahim, H. M., Carmody, T. J., Biggs, M. M., Suppes, T... & Kashner, T. M. (2004). The Inventory of Depressive Symptomatology, Clinician Rating (IDS-C) and Self-Report (IDS-SR), and the Quick Inventory of Depressive Symptomatology, Clinician Rating (QIDS-C) and Self-Report (QIDS-SR) in public sector patients with mood disorders: A psychometric evaluation. Psychological Medicine, 34(1), 73-82.

Whiston, S. C. (2009). Principles and applications of assessment in counseling (3rd ed.). Belmont, CA: Brooks/Cole, Cengage Learning.

Whiston, S., C. & Eder, K. (2003). Review of the BDI-FastScreen for Medical Patients. InB.S.Plake,J.C. Impara, & R.A. Spies (Eds.), The seventeenth mental measurements yearbook. Lincoln, NE: Buros Institute of Mental Measurements. Retrieved fromhttp://web.ebscohost.com.ezp.waldenulibrary.org/

Posted by at No comments:

Analyzing Tests and Scales




The following discussion will analyze the relationship among tests and scales, populations, and reliability and validity. In addition, it will describe some characteristics of the Hospital Anxiety and Depression Scale (HADS) and the Beck Depression Inventory Fast Screen for Medical Patients (BDI-FastScreen).

Scales are ordinarily constructed with the principle of unidimensionality which means the items that make up the scale, represent a single characteristic of a construct, and implementing the scale results in one number which represents the level of severity of the construct being assessed. For example, on the BDI-FastScreen, depression is rated between 0 and 21, representing no depressive symptoms and severe major depressive disorders respectively (Hennessey & Palone, 2003). The HADS rates users on a scale between 0 and 21, representing normal and severe anxiety/depression respectively (Campbell, 2007).

When social scientists construct scales and indexes, it is important that they assess validity and reliability, just as they would for any assessment. For example, the scale must be compared to other data that measures the same construct, contemporary research regarding the construct, and to other assessments of the same construct. The scales and indexes are correlated with other measures. These correlations attest to the instrument's validity. For example, the Hospital Anxiety and Depression Scale (HADS) was correlated to a 5-point psychiatric rating scales of anxiety and depression with 100 medical outpatients (Campbell, 2007). The Beck Depression Inventory Fast Screen for Medical Patients (BDI-FastScreen) was correlated with the Hospital Anxiety and Depression scales. Scales allow researchers to rank items by intensity, and changes ordinal and nominal data to interval level scales, which alleviates the limitations of nominal and ordinal data.

The samples upon which norms are based determines the generalizeability of the measure. A scale must be normed on a variety of populations to determine the extent the scale is reliable and valid in those populations. The BDI Fast-Screen was normed on 268 individuals drawn from four clinical environments. All were from the Philadelphia area. The norms for this assessment do not suggest generalizeability to other populations without further research (Hennessey & Pallone, 2003). That said, the BDI-FastScreen has been utilized with a variety of individuals, including patients with multiple sclerosis (Benedict, Fishman, McClellan, Bakshi, & Weinstock-Guttman, 2003), geriatric patients (Scheinthal, Steer, Giffin, & Beck, 2001), patients with chronic pain (Poole, Bramwell, & Murphy (2009) and cancer patients (Alacacıoğlu, Öztop, & Yılmaz, 2012). It may be safe to say that the BDI-FastScreen is generalizeable to a variety of medical patients, but not to individuals without medical conditions.

The authors of HADS list no information regarding norms in the manual, so it would be difficult to determine for which populations this scale would be appropriate, based on the manual alone. Additional research has been completed on adolescent populations, although without further research, this scale does not appear generalizeable to other populations. It has been developed to assess depression, anxiety, and irritability that results from medical conditions (Martin, 2007). This may be the only population for which the scale is reliable and valid.

References

Alacacıoğlu, A., Öztop, İ., & Yılmaz, U. (2012). The Effect of Anxiety and Depression on Quality of Life in Turkish Non Small Lung Cancer Patients. Turk Toraks Dergisi / Turkish Thoracic Journal, 13(2), 50-55. doi:10.5152/ttd.2012.12

Benedict, R., Fishman, I., McClellan, M., Bakshi, R., & Weinstock-Guttman, B. (2003). Validity of the Beck Depression Inventory-Fast Screen in multiple sclerosis. Multiple Sclerosis, 9(4), 393-396. doi: 10.1191/1352458503ms902oa

Campbell, M., H. (2007). Test review of the Hospital Anxiety and Depression Scale with theIrritability-Depression-Anxiety Scale and the Leeds Situational Anxiety Scale. In K. F. Geisinger, R. A. Spies, J. F. Carlson, & B. S. Plake (Eds.), The seventeenth mental measurements yearbook. Retrieved from the Burros Institute's Mental Measurements Yearbook online database.

Hennessy, J., J. & Pallone, N., J. (2003). Review of the BDI-FastScreen for Medical Patients. In B.S. Plake, J.C. Impara, & R.A. Spies (Eds.), The seventeenth mental measurements yearbook. Lincoln, NE: Buros Institute of Mental Measurements. Retrieved from http://web.ebscohost.com.ezp.waldenulibrary.org/

Martin, W., E. (2007). Test review of the Hospital Anxiety and Depression Scale with theIrritability-Depression-Anxiety Scale and the Leeds Situational Anxiety Scale. In K. F. Geisinger, R. A. Spies, J. F. Carlson, & B. S. Plake (Eds.), The seventeenth mental measurements yearbook. Retrieved from the Burros Institute's Mental Measurements Yearbook online database.

Poole, H., Bramwell, R., & Murphy, P. (2009). The utility of the Beck Depression Inventory Fast Screen (BDI-FS) in a pain clinic population. European Journal of Pain, 13(8), 865-869. doi: 10.1016/j.ejpain.2008.09.017

Scheinthal, S. M., Steer, R. A., Giffin, L., & Beck, A. T. (2001). Evaluating geriatric medical outpatients with the Beck Depression Inventory-FastScreen for Medical Patients. Aging & Mental Health, 5(2), 143-148. doi: 10.1080/13607860120038320

Whiston, S., C. & Eder, K. (2003). Review of the BDI-FastScreen for Medical Patients. In B.S. Plake, J.C. Impara, & R.A. Spies (Eds.), The seventeenth mental measurements yearbook. Lincoln, NE: Buros Institute of Mental Measurements. Retrieved from http://web.ebscohost.com.ezp.waldenulibrary.org/
Posted by at No comments:
Subscribe to: Posts (Atom)