Gardner-_ Hetznecker Sign and Symptom Check List

8/7/2019 Gardner-_ Hetznecker Sign and Symptom Check List

http://slidepdf.com/reader/full/gardner-hetznecker-sign-and-symptom-check-list 1/7

Journal of Gerontology1975, Vol. 30, No. 2, 209-215

Common Themes Among Moraleand Depression Scales

1

John N. Morris, PhD, Rosalie S. Wolf, BS,and Lorraine V. Klerman, DrPH 2

This paper reports on the infra- and interbattery scaling of three morale and depression batteriescomprised of self-reported items: PG C, G- H , and Z U N G . Responses to the three scales weresought from a sample of long-term residents of a state mental hospital. First, the batteries werefacto r analyzed sep arately to id entify the int raba ttery scales; then the resultant scales were comparedusing canonical correlation and super matrix factor analysis to identify the interbattery themes.These analyses resulted in the identification of a Clinical Depression domain across all three batteriesand single independent themes in both the Z U N G and PG C batteries.

H P H E assessment of mood states has received-*- considerable attention in both gerontology

and psychiatry. Gerontologists, concerned withmorale arid life satisfaction, have emphasizedvariations among individuals along dimensionsderived from hypotheses about theoreticallyand/or empirically relevant mood states (Law-ton, 1972; Neug arten, H avighurst, & Tobin,1961). Psychiatrists, concerned with clinical

depressed states, have dealt with perceptionsof distress, usin g self-re port mea sures to> cor-roborate <and enrich interview assessments(Beck, W ard, Mendelson, Mock, & Erbaugh ,1961; Lub in, 1966; Zun g, 1965 ).

The number of inventories and scales from

these two disciplines is large and growing.Salzman, Kochansky, Shader, and Cronin

(19 72) identified 39 mood inven tori es: 15 to

measure depression, 9 for morale, .and 15 for

other dimensions or more general phenomena.McNair (1973), in a recent survey of 75 trials

of the .effectiveness of antidepressant drugs,reviewed ten individual scales and three mis-cellaneo us categories of soales.

Self-report inventories are constructed on the

premise that some segment of a person's mood

state can be assessed through his or her responseto a structured battery of items. Although each

scale was developed from a specific frame of

reference and validated on a distinct population,redundancy in item content and structuring

* Presented at the Gerontological Society Annual M eeting, Miami

Beach, Nov. 9, 1973. This study was supported by the Brandeis-Worcester Tra ining P rogram in Social Research & Psychiatry, Tra in-ing Grant MH13154 (NIMH).

sFlorence Heller Graduate School for Advanced Studies in Social

Welfare, Brandeis Univ., W altha m, MA, and Worcester State Hos-pital, Worcester, MA.

is inevitable, particularly in the areas of moraleand depression. Since the inventories vary innumber and phrasing of questions, response al-ternatives, time frame, and other dimensions,it is difficult to compare study findings. Theexistence of many similar instruments, how-ever, suggests the desirability of interbatterycomparisons and scale reductions. This paperdescribes the extent of interbattery similarity

and uniqueness among one morale inventory(Philadelphia Geriatric Center Morale Scale)and two depression inventories (Gardner-Hetznecker Sign and Symptom Check List andZung Self-Rating Depression Scale).

The items included in these scales reflectthe discipline of their creators, gerontology or

psychiatry. Lawton (1972) indicated that the

PGC battery of items was designed to sort

"out the components of morale."In. the definition used . . . the components . . .stressed are freedom from distressing symptoms,satisfaction with self, feeling of syntony between self

and environmenit, and ability to strive appropriatelywhile still accepting the inevitable. An effort wasalso made to anchor within the definition, some ofthe characteristics which have been considered lesscentral to morale.

Zung (1972), however, stated that his depres-

sion battery was. . . construdted on the basis of the clinical diagnosticcriteria most commonly used to characterize depres-sion disorders in terms of the presence of a pervasiveaffective mood of feeling depressed, with concomitantphysiological and psychological disturbances.

S T U D Y P O P U L A T I O N

When first interviewed, the study subjects

were patients in a state mental hospital that

was being phased out. The philosophy of care

209



210 MORRIS, WOLF, AND KLERMAN

within the institution and the cessation ofadmissions during the phase-out led to a largely-chronic residual population. In late 1972 120

patients were selected for transfer to a nearbystate hospital and they comprise the study groupon which this paper is based. Of these 120patients, 89 responded to the self-report ques-tions within the three morale-depression in-ventories during either the Time 1 (N=77) orthe Time 2 ( N = 7 2) interview, 15 weeksapart. The median age of these 89 patients was53 years . with a range of 20 to 86 years.The male-female ratio was 51% to 49%,and the black-white ratio was 6% to94%. Most were long-term state hospitalpatients: 99% had been hospitalized for 2 ormore years, and 55% for over 9 years.

PROCEDURES

The three measures of mood state, the Phil-adelphia Geriatric Center Morale Scale (PGC),the Gardner-Hetznecker Sign and SymptomCheck List (G-H), and the Zung Self-RatingDepression Scale (ZUNG), were selected onthe basis of their reputation in their respectivefields as well as their conceptual and structuraldifferences. The PGC (Lawton, 1972) wasdeveloped to measure morale in the elderlyand the ZUNG (Zung, 1965) was addressedto the question of depression in a more generalbut depressed population. The G-H (Hetz-necker, Gardner, Odoroff, & Turner, 1966), aless well-known depression battery, whilesimilar in content to both PGC and ZUNG,differs in structure since it does not considerresponse-set biases, i.e., both the 20-itemZUNG and the 22-item PGC batteries shiftthe wording of items between positive andnegative poles, while the 17-item G-H is pre-sented within a totally negative framework.

The PGC scale was scored according to Law-ton, i.,e.., all items were dichotomized exceptone, which was reduced to a dichotomy, butwas limited to the 17 items recently suggestedby Morris and Sherwood (1975). The ZUNGand G-H batteries were reduced from the sug-gested four-point scales to three-point scales(never, or almost never; sometimes; often);and one item was omitted from each batterywhen they were administered to the studysubjects. In the case of ZUNG, an item con-

cerning sexual interest was dropped as suggestedby Salzman et al. (1972). One G-H item, how-ever, was inadvertently omitted from the sched-ule and should have been included.

In this type of study, the investigators arelimited by the reluctance of subjects to repeat-edly respond to similar questions and a balance

must be struck between research* [email protected] rights. The present study interspersedthe three batteries of morale and depr^ssiosaitems within a larger 1-hour interview scheduleseparated by questions aimed at other dimen-sions of the patient's experience. The trainedinterviewer (either a social worker or sociol-ogist) read both the question and the specifiedresponse alternatives, and the respondent wasasked to confine his reply accordingly.

The resultant individual item responsestended toward "J" shaped distributions, al-

though many items had more flattened distribu-tions. Only one item had over 80% of theresponses within a single category.

STATISTICAL ANA LYSES

The statistical comparison of the threemorale-depression inventories was conducted in

five stages, with each succeeding step building

on the previous one.(1) Each of the three inventories of items was

factored separately to reveal its fundamental structure,and scales were developed based on items with highloadings for both Times 1 and 2. The total pool ofitems from the three inventories was not factored atthis point because of possible presence of theoreticallynon-congruent or mathematically inappropriate items.

(2) The scales from Step 1 were compared on apair-wise basis using canonical correlation to identifyinterscale similarities, as well as dissimilarities.

(3) The total pool of scales from Step 1 were alsocompared simultaneously using super matrix correlationto detect similar conceptual threads.

(4) Items from those scales found to be interrelatedin Steps 2 and 3 were factored to uncover the strongestset representing the common theme. This set of itemswas identified as a new cross-inventory scale.

(5) Finally, the Step 1 scales and the cross-inventoryscale from Step 4 were intercorrelated in their raw-

score summated formats to demonstrate the extent towhich they are statistically independent.

Factor Analysis of Each Inventory

Each of the three inventories was factoredseparately for Time 1 and 2 using the methodof Principal Components followed by VarimaxRotation. The rotated factors involving thefirst two through six principal components wereinspected for solutions both consistent acrossthe two time periods and subject to meaningfulinterpretation.

In the G-H battery (Table 1) Factor 2,Dejection, a sense of emptiness, hopelessness,and deprecation, was the most similar at bothtimes. Factors 1 and 3 were less clear, since



MORALE AND DEPRESSION THEMES 211

some items loaded at one t ime but not at the tions are followed, indicating two factors: Tran-other. Nevertheless, because major anch or items quility, or its opposite, subjective discomfort,appeared at each time, indicating a similarity consisting of items measuring fear, control, and

in theme, if not in total structu re, both factors life acceptance; and Life Progression, consist-were accepted. Factor 1 was labelled Impaired ing of measures of awareness of aging or lifeThinking, difficulty in organizing thought% in changes over time.

concentration, and in ridding mind of unpleas- The Z UNG (T a b le 3) analysis resulted in

an t or nonsensical thoughts; and Factor 3, the clearest rotated factor solution. Factor 1,

Gu ilt. labeled Agitation, contained items of irritability,The factors for the PGC battery (Table 2), restlessness, and despair. Factor 2, Self-satis-

while not totally consistent, resemble the con- faction, included measures of personal enjoy-

figuration previously reported by Morris and ment, usefulness, and hopefulness.Sherwood (1975). Their suggested item selec- The presence of a multidimensional structur-

Table 1. Varimax Rotation of Three Principal Component Factors for G-H Battery of Items.

Factor LoadingsImpaired Thinking Dejection Guilt

Time 1" Time 2b Time 1 Time 2 Time 1 Time 2

Wants to cry but can't

Fears something unpleasant is going to happen

Has terrifying, anxiety-provoking dreams

Feels depressed (sad, gloomy, hopeless, low in spirit)

Future empty, bleak, no plans

Feels himself a failure, deprecates himself

Feels he causes suffering or harm to others

Feels he deserves punish ment, feels guilt

Has trouble ridding mind of unpleasant or nonsensical though ts

More than usual amount ofdaydreaming

Unable to concentrate on tasks at hand

Has difficulty in keeping thoughts organizedFeels distant, isolated from things and people

Prefers being alone, seldom feels comfortable with anyone

Fears he is losing his mind

No one understands him

Latent Root

• time 1 n = 77.b

time 2 n = 72. *Items defining the factor.

Table 2. Varimax Rotation of Principal Components51

for PGC Bat tery of I tems.

Factor Loadings

Tran quil ity Life Progression

^ Time 1 Time 2 Time*1! Time 2

Things keep getting worse as I get older

I have as much pep as I did last year

Little things bother me more, this year

As you get older you are less useful

I sometimes worry so much tha t I can' t sleep

As I get older, things are better , worse or same

I sometimes feel that life isn't worth living

I am as h a p p y now as I was when I was younger

I have a lot to be sad about

People had it be t te r in the old days

I am afraid of a lot of th ings

I get mad more than I used to

Life is hard for me most of the t ime

How satisfied are you with your life today? (satisfied, no t satisfied)

I take things hard

A person has to live for today and not worry about tomorrow

I get upset easily

La tent Root

» Time 1 Based on 4 rotated factors. Time 2 Based on 3 rota ted factors. *Item s defining th e factor.

0 .140

- 0 . 0 5 3

0.215

0.443

0.336

0.239

0.094

0.166

•0.472

•0.628

•0.806

•0.8170.281

0.170

0.228

-0 .294

2.652

-0 .0 1 9

0.367

0.747

0.126

0.288

0.380

0.522

0.669

0.462

0.824

0.214

-0 .0400.505

0.290

0.664

0.337

• 3.492

• -0 .547

• -0 .690

• -0 .516

• -0 .565

• -0 .632

• -0 .529

-0 .422

-0 .290

-0 .205

-0 .173

-0 .098

-0 .123• -0 .540

0.088

-0 .1 0 3

• -0 .546

3.007

-0 .6 7 5

-0 .4 5 5

-0 .2 1 9

-0 .800

-0 .7 6 9

-0 .6 0 8

-0 .2 6 6

-0 .3 8 1

-0 .2 9 3

0.062

-0 .3 8 0

-0 .3 4 1-0 .4 4 2

0.122

-0 .3 5 0

-0 .5 7 6

3.540

-0 .1 0 8

0.294

0.088

0.006

0.175

0.373

•0.695

•0.729

0.234

0.408

0.240

0.087

•0.402

•0.705

•0.672

0.228

2.738

-0 .4 2 4

-0 .5 2 6

-0 .0 3 7

-0 .1 1 5

-0.C2O

-0 .3 2 3

-0 .6 9 1

-0 .1 9 8

-0 .5 0 1

-0 .2 1 5

-0 .6 8 4

-0 .6 0 8-0 .3 2 5

-0 .696

-0 .3 5 1

-0 .3 0 3

3.031

0.2.51

0.136

•0.4340.352

•0.714

-0 .0 1 1

0.424

-0 .0 7 9

•0.393

0.054

•0.736

•0.681

•0.494

•0.279

•0.775

0.062•0.725

3.705

0.423

0.199

0.609

0.290

0.607

0.189

0.474

-0 .0 0 1

0.616

-0 .0 1 0

0.759

0.640

0.730

0.492

0.677

-0 .1 5 40.699

4.413

•-0.649

•-0.582

•-0.603

•-0.561

-0.179

•-0.511

•-0.365

•-0.184

-0 .1 9 7

-0 .2 0 2

-0 .1 8 4

-0 .2 8 8

-0 .3 0 7

0.098

0.017

-0 .0 9 2-0 .2 8 2

2.286

-0 .4 6 2

0.508

0.027

-0 .5 6 1

-0 .1 9 2

-0 .3 2 6

-0 .3 8 5

-0 .0 1 1

-0 .3 3 4

-0 .7 4 8

-0 .3 2 5

-0 .0 3 7

-0 .1 8 9

0.205

0.032

0.103-0 .0 4 7

1.948




Table 3. Varimax Rotation of Principal Components for Zung Battery of Items.

I feel downhearted and blue

Morning is when I feel the best

I have crying spells or feel like it

I have trouble sleeping at night

I eat as much as I used to

I notice that I am losing weight

I have trouble with constipation

My heart beats faster than usual

I get tired for no reason

My mind is as clear as it used to be

I find it easy to do the things I used to

I am estless and c an't keep still

I feel hopeful about the future

I am more irritable than usualI find it easy to make decisions

I feel that I am useful and needed

My life is pretty full

I feel that others would be better off if I were dead

I still enjoy the things I used to do

Latent Root

Time 1

•0.680-0 .048

•0.738

•0.599

-0 .072

0.210

•0.674

•0.513

•0.702

0.012

0.007

•0.714

0.154

•0.6810.130

0.158

0.067

•0.622

0.068

4.062

Factor Loadings

Agitation

Time 2

0.622

-0 .437

0.773

0.564

-0 .117

0.547

0.700

0.683

0.675

0.033

-0 .0 2 1

0.629

0.036

0.7620.045

0.095

0.082

0.723

0.257

4.804

Self-Satisfaction

Timel

0.1490.375

0.087

0.195

•0.431

-0.159

0.031

-0 .014

0.093

•0.570

•0.734

0.077

•0.705

-0 .205•0.637

•0.701

•0.711

0.044

•0.783

3.855

Time 2

0.353

0.299

0.012

0.261

0.375

-0 .0 8 3

-0 .159

-0.059

0.184

0.495

0.774

0.244

0.597

-0 .0 3 30.469

0.751

0.749

0.167

0.707

3.625

•I tems defining the factor .

ing for each of the scale batteries was sub-stantiated further by K.R. 20 Alpha Reliabilitycoefficients at Times 1 and 2. These values(Table 4) indicated the extent to which items

used in interpreting the several scales (indicatedby the asterisks in Tables 1, 2, & 3) were statis-tically dependent when summated in their rawscore format. In an exploratory study, such asthe present one, scales may be accepted as re-liable if they attain a K.R. 20 value of .50 orgreater (Nunnally, 1967). Each of these sevenscales exceeded this value at both Times 1 and2, and most of the coefficients were considerablyhigher.

Interscale Structuring—Canonical Correlation

Canonical correlational procedures were usedto provide a measure of the pair-wise cross-inventory interrelationships between the scalesgenerated within each of the three morale-de-pression batteries. To achieve this end, thecanonical procedures generate prediction equa-tions that determine the linear combinationsof the items in one battery that are maximallyrelated to the linear combination of items inthe second battery. Canonioal correlation canbe seen as

. . . essentially a procedure for factoring two batteriessimultaneously, in order to extract factors which areuncorrelated within their batteries but which providemaximum correlations of pairs of factors across bat-teries. That is, the first factor of each battery is

Tab le 4. KR -20 Alpha R eliabilities for Scales Identifiedthrough Factor Rotations.

GH — Impa ired Thinking

GH — Dejection

G H — G u i l t

ZUNG — Agi ta t ion

ZU NG — Self-Satisfaction

PGC — Tranqui l i ty

PGC — Life Progression

T i m e l

.766

.774

.748

.851

.827

.844

.714

Time 2

.686

.876

.818

.870

.785

.848

.690

located so th at the canonical correlation of the firstfactor is maximized. The resulting coefficient is thelargest product-moment correlation that can bedeveloped between linear functions of the two bat-teries. It expresses the maximum redundancy of apair of factors, one from each of the batteries.(Cooley & Lohnes).

Within the present body of data, the canon-ical correlation of the three G-H scales and thetwo PGC scales (at both Times 1 and 2) wasgreater than .72 and significant at less than the.001 level. Based on the canonical loadings(which are interpreted in the same way asfactor loadings) from this comparison, all threeof the G-H scales—Dejection, Impaired Think-ing, and Guilt—were significantly related toboth of the PGC scales, Tranquility and LifeProgression. The level of redundancy between

the G-H and ZUNG scales was slightly higher(.75 at Time 1 and .87 at Time 2) and signi-ficant at less than the .001 level, but in thisinstance the three G-H scales were related to




only one of the two ZUNG scales, Agitation'The remaining ZUNG scale, Self-satisfaction,was largely independent of the G-H scales.

When the ZUNG scales were compared withthe PGC scales, a similar finding emerged: firstcanonical (.80 here at Time 1 and .79 at Time 2)significantly (.001 at both time periods) relatedthe PGC Tranquility and Life Progression scalesto the ZUNG Agitation soale only. A secondsignificant (.01 at both time periods), butweaker, relationship emerged between ZUNGSelf-satisfaction and PGC Life Progression (asecond canonical of .31 at Tim e 1 and .29 atT ime 2 ) .

These data provide a first indication of the

considerable informational or conceptual over-lap between the constructed scales from thethree morale-depression inventories (the de-tailed substantiating tables .are available fromthe senior author upon request). A more de-tailed exposition of these functional relation-ships is developed in the following super matrixfactor analyses.

Interscade Structuring—Super Matrix Factor Analysis

The findings from the pair-wise canonical

correlational analyses were substantiated andclarified through the application of super matrix

factor analysis or simultaneous canonical cor-relation (Horst, 1961a, b), a procedure for

simultaneously considering the interrelation-

ships among the scales from all three morale-

depression inventories. This procedure is muchlike oanonical correlation, only in this instance

the procedure generates the maximum correla-

tion between linear functions of the scales fromall three inventories simultaneously. The

procedure, therefore, can be seen as involving

the com putation of principal components withineach inventory and subsequently intercorrelat-ing the resultant principal component scores of

the sample members across the three inven-

tories. The simultaneous correlation is a meas-ure of the general overlap among the three in-

ventories, and. . . will always be equal to or less than the pair-wise maximum canonical correlation. This is truebecause simultaneous canonical correlation maximizes,general overlap rather than pair-wise relationships(Jones & Jones, 1970).

For substantial simultaneous canonioals, the

final interpretation falls upon the individualscale loadings, which are interpretable within

the same frame of reference as factor loadings,

or the more normal canonical loadings, andare the correlation of the item with a con-structed score.

Using this method, the first two roots fromboth points in time summarized th e informationoverlap between the three batteries. Root onewas substantial (.68 and .75 respectively) andcaptured most of the G-H items, the PGCTranquility items, and to a lesser extent, thePGC Life Progression items and the ZUNGAgitation items. Root two, which was not large(.18 and .29, respectively), included the ZUNGSelf-satisfaction items and the PGC Life Pro-gression items—although the latter is true forTim e 1 only. This .analysis clearly de mon strated

the existence of at least two themes across thesethree batteries, with the possibility that thePGC Life Progression scale might be a thirdindependent component.

Factor Analysis of Derived Scales

The next step in the search for interbatterythemes was to j'ointly factor .analyze items fromthose scales shown to be similar by the twocanonical correlation procedures. The items inthe ZUNG Self-satisfaction scale were omittedfrom these computations, since evidence from

the Super Matrix, canonical correlation, as wellas the Varimax rotations, had indicated thatthey were in an independent domain.

The first principal component solution foritems in the remaining six scales yielded con-sistently high loadings, with the exception offour of the six measures making up the PGCLife Progression soale. On the basis of this find-ing and the partial relationship of this scalewith the ZUNG Self-satisfaction scale in thesuper matrix and canonical correlation, the sixitems in the PGC Life Progression scale were

dropped from the pool. Thus, both the PGCLife Progression and th e Z U N G Self-satisfactionscales were designated "ind epend ent" themes.

The items from the remaining five scaleswer e factored agai n ( Ta bl e 5) . Five, items from

the G-H battery, three from the PGC, and

four from the ZUNG had a loading of .6 orhigher on the first principal component at both

Times 1 and 2. This comm on theme was labeled

Clinical D epression, since the items were similarto those used by a trained clinician in making

a diagnosis of depression (the 12 items of the

Clinical Depression soale are indicated byasterisks on Table 5). K.R. 20 Alpha Reliabilitycomputation based on these items resulted in a




Time 1 coefficient of .891 and a Time 2 of .908,indicating the presence of a highly internallyconsistent scale.

Table 5. Factor Loadings for First Principal Com ponentfor all I tems in Following Scales.

G-H

G-H

G-H

Impaired Thinking

Dejection

Guilt

PGC Tranq ui li ty ZUNG Ag itat io n

Time 1 Time 2

G-H Wants to cry but can't 0.376 0.624

G-H Fears something unpleasant is going to happen 0.559 0.758

G-H Has terrifying, anxiety-provoking dreams 0.433 0.523

G-H Feels depressed, (sad, gloomy, hopeless, down

in dumps) 0.500 0.582

G-H Futur e empty, bleak, no plans 0.611 0.574

•G-H Feels himself a failure 0.606 0.705

*G-H Feels he causes suffering or harm to others 0.674 0.805

*G-H Feels he deserves punishment, feels guilt 0.642 0.693G-H Has trouble ridding mind of unpleasant

thoughts 0.444 0.636

G-H More than usual amount ofdaydreaming 0.643 0.507

G-H Unable to concentrate on tasks at hand 0.578 0.752

G-H Has difficulty in keeping thoughts organized 0.514 0.540

*G-H Feels distant, isolated.from things and people 0.667 0.721

G-H Prefers being alone, seldom feels comfortable

with anyone 0.352 0.483

•G-H Fears he is losing his mind 0.623 0.759

G-H No one understands him 0.265 0.686

PGC Little things bother me more this year 0.519 0.481

PGC I sometimes worry so much th at I can't sleep 0.519 0.595

PGC I have a lot to be sad about 0.531 0.594

•PGC I am afraid of a lot of things 0.638 0.656

*PGC I get mad more than I used to 0.753 0.598

•PGC Life is hard for me most of the time 0.602 0.611

PG C How satisfied are you with your life today?

(satisfied, not satisfied) 0.412 0.456

PGC I take things hard 0.733 0.511

PG C I get upset easily 0.687 0.514

•ZUNG I feel dow nhearted and blue 0.690 0.722

•ZUNG I have crying spells or feel like it 0.677 0.753

ZUNG I have trouble sleeping at night 0.540 0.655

ZUNG I have trouble with constipation 0.466 0.631

ZUNG My heart beats faster than usual 0.452 0.494

ZUNG I get tired forno reason 0.628 0.587

•ZUNG I am restless and can't keep still 0.633 0.622

ZUNG I am more irritable than usual 0.565 0.687

•ZUNG I feel tha t others would be better off if

I were dead 0.646 0.653

Latent roo 11.250 13.471

•Items defining Clinical Depression scale.

Table 6. Time 1 and Time 2 Scale Distribution for:Clinical Depression.

Score Value

Timel

N %

Time 2

0 - 2

3 - 5

6 -8

9-11

12-14

15-17

18-20

21-23

24

27

19

10

8

3

4

3

2

1

35.0

24.7

13.0

10.4

3.9

5.2

3.9

2. 6

1.3

X = 6.117

Standard D eviation = 6.152

Possible Range = 0-24

27

14

9

8

4

2

4

1

3

6.556

6.747

0 - 2 4

37.5

19.4

12.5

11.2

5.5

2.8

5.5

1.4

4.2

Intertheme Correlations

Zero-order correlations were computed forthe three scales within their raw score sum-

mated formats. The shared variance (the zero-order correlation squared) at Time 1 betweenthe Self-satisfaction and Life Progression scaleswas 25%, between Self-satisfaction and ClinicalDepression, 8%, and between Life Progressionand Clinical Depression, 36%. These relation-ships hold at both Times 1 and 2. The presenceof the low levels of soale overlap indicates thatthe three internally consistent raw score sum-mated versions of the scales capture uniquecomponents of the individual's inner moodstate. The three themes of Clinical Depression,

Self-satisfaction, and Life Progression, there-fore, are both mathematically distinct and con-ceptually cohesive.

Sample Scale D istributions

The research team wished to know the dis-tribution of the scores on the three independentscales within the study sample (Tables 6-8).The patients tended to be less depressed, ormore satisfied, in the two presently orientedscales—Clinioal Depression (Table 6) andZU NG Self-satisfaction (Table 7)— and to

show a negative orientation towards perceivedlife changes over time in the longitudinallybased PGC Life Progression scale (Table 8).

Table 7. Time 1 and Time 2 Scale D istribution for:ZUNG Self-Satisfaction.

T im e l Time 2

Score Value

0 - 2

3 - 5

6 - 8

9 - 1 1

1 2 - 1 41 5 - 1 6

28.5

19.5

18.2

22.1

9.12. 6

26

25

7

9

5

4

3

0

.375

.777

- I f i

36.2

34.7

9.7

12.5

6.9

X = 6.013

Standard Deviation = 4.275

Possible Range = 0-16

Table 8. Time 1 and Time 2 Scale Distribution for:PG C - Life Progression

Time l Time 2

Score Value

0 - 2

3 - 5

6 - 8

9 - 1 1

1 2 - 1 4

XStandard D eviation

Possible Range

27

12

20

10

8

=

=

=

5.

3.

0

558

982

-1 4

35.0

15.6

26.0

13.0

10.4

24

8

23

11

6

5.

3.

0

556

876

- 1 4

33.3

11.2

31.9

15.3

8.3




The ZUNG Self-satisfaction scale (Table 7)shows a significant Time 1 to Time 2 shift inmeans, suggesting on this dimension, but not on

the other two, an increase in satisfaction atTime 2, 15 weeks after the initial data gather-ing. This may reflect *a perceived benefit result- .ing from the move from one state hospital toanother. The failure of this shift to be mirroredin all three scales is a further indication of theindependence of the scales and suggests thatthese dimensions may be affected by uniquesets of oasual forces.

SUMMARY

Numerous studies have attempted to assess

morale and depression. The instruments, how-ever, often have reflected the researcher's ori-entation, making it difficult to determine thesimilarity in the. intrapsy chic states being meas-ured. This -paper has addressed the problemdirectly by seeking common themes amongthree scales selected from the gerontologicaland psychiatric fields: the Philadelphia Geria-tric Center Morale Scale (PGC), the Gardner-Hetznecker Sign and Symptom Check List(G-H), and the Zung Self-Rating DepressionScale (ZUNG). Seven sets of items were pro-

duced through factor analyses, two of whichrepresented independent themes, the ZungSelf-satisfaction (asterisk items in Table 3) andthe PGC Life Progression (asterisk items inTable 2). The other five focused on thoseaspects of morale and depression associated withpsychic distress. These were determined to bemathematically similar and the construct waslabeled Clinical Depression (asterisk items inTable 5). It included .elements of the G-HImpaired Thinking, Dejection, and Guilt, thePGC Tranquility, and the ZUNG Agitation

scales (asterisk items in Tables 1, 2, and 3).The existence of this common theme across

independently constructed scales suggests thatdepression and low morale, to some degree,represent an equivalent mood. The pooled scaleitems not only provide a more reliable measureof this state than the separate inventories, butconstitute a more valid instrument for crossdisciplinary studies. Any of the five scales fallingwithin the Clinical Depression domain, how-ever, could have been selected as the one meas-ure of the common mood state. Within the

gerontological field, where the PGC battery ismore prevalent, the PGC scale componentmight be designated the Clinical Depression in-

dicator in either the Morris and Sherwood(1975) format or in one of the two formatssuggested' by Lawton (1975). An i importantand independent dimension would be, addedif the ZUNG Self-satisfaction domain also wereadministered.

REFERENCES

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., &Erbaugh, J. An inventory for measuring depression.Archives of General Psychiatry, 1961, 4, 561-571.

. Cooley, W., & Lohnes, P. Multivariate data analysis.John Wiley & Sons, New York, 1971.

Hetznecker, W., Gardner, E., Odoroff, C. L., & Turner,R. J. Field survey methods in psychiatry. Archives ofGeneral Psychiatry, 1966,15, 427-438.

Horst, P. Generalized canonical correlations and theirapplication to experimental data. Journal of-ClinicalPsychology, 1961, Monograph Supplement 14, 615-620.

• (a )

Horst, P. Relations among m sets of measures. Psy-chometrika, 1961, 26, 129-149. (b)

Jones, K. J., & Jones, P. P. Contributions of the Ror-schach to description of personality structure dennedby several objective tests. Psychological Reports, 1970,26, 35-45.

Lawton, M. P. The dimensions of morale. In D. P.

Kent, R. Kastenbaum, & S. Sherwood (Eds.), Re-search planning and action for the elderly.BehavioralPublications, New York, 1972.

Lawton, M. P. The Philadelphia Geriatric Centermorale scale: A revision. Journal of Gerontology,1975, 30, 85-89.

Lubin, B. Fourteen brief depression adjective check-lists. Archives of General Psychiatry, 1966, 25, 205-208.

McNair, D. M. Self-evaluations of anti-depressants.Prepared for the FDA-ACNP Task Force Subcom-mittee on Anti-depressanits. Rev. Aug., 1973.

Morris, J. N., & Sherwood, S. A retesting and modifica-tion of the Philadelphia Geriatric Center moralescale. Journal of Gerontology, 1975, 30, 77-84..

Neugarten, B. L., Havighurst, R. J., & Tobin, S. S.The measurement of life satisfaction. Journal ofGerontology, 1961, 16, 134-143.

Nunnally, J. C. Psychometric theory. McGraw-Hill,New York, 1967.

Salzman, C , Kochansky, G. E., Shader, R. I., & Cronin,D. M. Rating scales for psychotropic drug research ingeriatric patients. II. Mood ratings. Journal of theAmerican Geriatrics Society, 1972, 20, 215-221.

Zung, W. A self-rating depression scale. Archives ofGeneral Psychiatry, 1965, 12, 63-70.

Zung, W. W. K. How normal is depression? Psycho-somatics, 1972, May-June, 23, 27-31. .

Gardner-_ Hetznecker Sign and Symptom Check List

Documents