Sentence Completion Test

1. Abstract

The Sentence Completion Test developed by Herman Y. Efron in 1960 represents an early, historically significant attempt to employ semi-projective verbal methodology for clinical risk stratification in psychiatric hospital settings. Specifically designed as an objective-projective screening instrument, the instrument aims to identify psychiatric inpatients manifesting active or latent suicidal ideation, as well as to differentiate suicidal tendencies from outwardly directed assaultive behavior and non-suicidal/non-assaultive psychiatric controls. The test comprises 54 incomplete sentence stems, strategically blending stems adapted from antecedent sentence completion batteries with newly constructed stems explicitly targeting depressive symptomatology, despair, morbid preoccupation, interpersonal friction, and self-evaluative attitudes. Unlike purely subjective projective assessment paradigms that rely on qualitative psychoanalytic interpretation, Efron’s paradigm instituted an empirical classification framework wherein clinical raters evaluated the completed protocols to categorize respondents into one of three clinical cohorts: suicidal, assaultive, or non-suicidal/non-assaultive, accompanied by a quantitative rating of judge confidence. Psychometric investigations conducted during its development revealed nuanced challenges characteristic of early projective and objective screening efforts in clinical psychology. Specifically, while the instrument facilitated qualitative clinical insights into the patient’s phenomenological world, empirical indices of inter-rater reliability, classification accuracy, and diagnostic validity highlighted the profound complexity of predicting low-base-rate behaviors such as suicide using semi-structured verbal protocols. Modern psychometric evaluations emphasize that while the 54-item instrument lacks contemporary confirmatory factor-analytic substantiation and standardized normative tables, it occupies an essential conceptual bridge in the evolution of clinical suicidology, illustrating the methodological transition from unstructured psychodynamic inference toward objective screening criteria and standardized behavioral classification.

2. Keywords

Sentence Completion Test, Herman Y. Efron, Suicidal Ideation, Psychiatric Inpatients, Projective Techniques, Suicide Risk Screening, Assaultive Behavior, Clinical Judgment, Semistructured Measures, Depressive Themes, Inpatient Assessment

3. Authors

The instrument was developed by Herman Y. Efron, Ph.D., who served as a clinical research psychologist affiliated with the Veterans Administration (VA) healthcare system during the mid-twentieth century. His primary work was conducted at the Veterans Administration Hospital in Lyons, New Jersey, United States. Efron was an active contributor to clinical psychology literature during the late 1950s and 1960s, focusing heavily on experimental psychopathology, institutional psychiatric evaluation, objective scoring of projective techniques, and the clinical prediction of self-directed and other-directed violence among hospitalized psychiatric patients.

4. Purpose

The primary purpose of the Sentence Completion Test developed by Efron (1960) was to provide a systematic, group-administrable screening device capable of detecting covert or overt suicidal ideation among newly admitted or hospitalized psychiatric patients. In inpatient psychiatric environments during the mid-20th century, clinicians faced an urgent diagnostic imperative: identifying patients at acute risk for self-harm before such impulses manifested in irreversible action, while simultaneously differentiating self-destructive impulses from outwardly aggressive or assaultive tendencies.

Direct clinical inquiry often proved insufficient, as patients frequently engaged in impression management, denial, or cognitive evasion due to institutional stigma or fear of prolonged involuntary confinement. Efron recognized that standard direct questionnaires could be easily manipulated by guarded patients, whereas open-ended expressive projective instruments (such as the Rorschach Inkblot Test or Thematic Apperception Test) were labor-intensive, time-consuming, and difficult to quantify reliably across large hospital admissions. Consequently, Efron’s test was constructed to occupy an intermediate psychometric space—a semi-structured verbal instrument that maintained the projective power of expressive completion while enabling rapid, standardized paper-and-pencil administration to groups of patients upon admission.

Beyond detecting self-destructive thoughts, the theoretical and clinical rationale was explicitly differential. The measure aimed to parse two distinct manifestations of behavioral disruption in psychiatric facilities: self-directed hostility (suicide) versus other-directed hostility (assaultiveness). By presenting patients with open-ended sentence stems heavily saturated with themes of depression, defeat, frustration, and existential loss, the test aimed to elicit distinctive cognitive and emotional reaction patterns. In clinical research applications, the instrument served as a platform to investigate the diagnostic accuracy and confidence calibration of clinical judges, examining whether professional psychologists and psychiatric trainees could reliably infer life-threatening behavioral trajectories from semi-projective verbal products.

5. Psychological Construct

The core psychological construct measured by Efron’s instrument is Suicidal Ideation in Hospitalized Psychiatric Patients, conceptualized through verbal manifestations of self-destructive intent, depressive despair, perceived helplessness, and hostility directed inward. In the operationalization of this measure, suicidal ideation is not treated merely as an isolated affirmative symptom, but rather as an overarching constellation of affective, cognitive, and dynamic orientations that can be contrasted against outwardly aggressive pathology.

Although the 54-item test was administered as a continuous semi-projective protocol rather than divided into formally partitioned numerical subscales, its clinical scoring and interpretive framework encompass three primary latent thematic dimensions:

• Depressive Affect and Existential Hopelessness: This dimension is elicited by sentence stems designed to tap the respondent’s cognitive representations of the future, personal worth, and endurance under psychological distress. Stems addressing emotional gloom, fatigue, perceived pointlessness of effort, and despair serve as projective stimuli. Patients harboring latent suicidal intent are hypothesized to project deep hopelessness, pervasive helplessness, and a longing for relief or cessation of consciousness into their grammatical completions.
• Hostility Direction and Aggression Orientation (Inward vs. Outward): Grounded in classical psychoanalytic and behavioral conceptions of violence, this construct examines whether latent aggression is channelled inwardly against the self (suicide) or outwardly against the environment (assaultiveness). Responses indicating personal culpability, profound self-hatred, and guilt reflect inward-directed aggression, whereas externalized blame, expressions of rage toward authority figures, and violent intent toward peers indicate outward assaultiveness.
• Interpersonal Alienation and Institutional Maladjustment: This dimension captures the respondent’s subjective perception of social connection, therapeutic alliance, and institutional containment. Stems probing relationships with hospital staff, family members, and society evaluate whether the patient experiences supportive belonging or insurmountable alienation, an established catalyst for acute self-destructive crises.

6. Theoretical Framework

The theoretical architecture of Efron’s Sentence Completion Test is historically anchored at the intersection of psychodynamic theory, early cognitive theories of depression, and the mid-century projective movement. Psychodynamically, the instrument is indebted to Sigmund Freud’s foundational formulation in Mourning and Melancholia (1917), which conceptualized melancholia and suicidal ideation as the consequence of hostility originally aimed at an ambivalently loved lost object that has been turned inward against the ego. Under this psychoanalytic premise, suicide represents the ultimate act of retroflected rage. Furthermore, Karl Menninger’s seminal thesis in Man Against Himself (1938) posits that self-destruction comprises three distinct components: the wish to kill, the wish to be killed, and the wish to die. Efron’s sentence stems were tailored to provoke these unconscious or semi-conscious dynamic drives by confronting the patient with projective cues related to life, failure, punishment, and hostility.

Simultaneously, the test reflects the principles of projective methodology established by Julian Rotter and others who developed sentence completion techniques during the 1940s and 1950s (such as the Rotter Incomplete Sentences Blank). The projective hypothesis assumes that when individuals are confronted with ambiguous, incomplete linguistic stimuli, they spontaneously project their idiosyncratic needs, conflicts, defense mechanisms, and preoccupations into their completions. Unlike fully unstructured stimuli, sentence stems provide partial structural constraints, guiding the patient’s focus toward specific emotional conflicts (such as depression and interpersonal hostility) while allowing freedom in how the final thought is organized and resolved.

Finally, the instrument anticipated cognitive formulations of suicide later consolidated by Aaron T. Beck in his cognitive triad of depression. Efron’s inclusion of heavily depressive stems was based on the premise that suicidal individuals exhibit systematic cognitive biases—pervasive negative views of the self, the world, and the future. By presenting open sentence fragments, the test prompts the patient to articulate these underlying cognitive schemas, providing empirical clinicians with observable verbal markers of cognitive constriction and hopelessness.

7. Validity

Empirical evaluation of the construct, predictive, convergent, and discriminant validity of the 1960 Sentence Completion Test was documented in Efron’s original empirical investigation published in the Journal of Consulting Psychology. The methodological design centered on evaluating clinical validity via a criterion-group differentiation paradigm rather than formal psychometric correlation matrices.

To assess diagnostic and predictive validity, Efron administered the 54-item instrument to three distinct cohorts of male psychiatric inpatients: patients with documented active suicidal thoughts or behavior, patients with documented histories of assaultive or violent institutional behavior, and psychiatric control patients exhibiting neither suicidal nor assaultive histories. Clinical judges (including experienced clinical psychologists and clinical psychology trainees) were presented with the blind transcripts of the completed sentence tests and instructed to classify each patient into one of the three diagnostic categories (suicidal, assaultive, or non-suicidal/non-assaultive).

The empirical results yielded important insights into the limitations of intuitive clinical judgment based on projective completions:

• Diagnostic Classification Accuracy: Clinical judges demonstrated classification accuracy that barely exceeded chance levels when attempting to differentiate suicidal patients from assaultive or non-suicidal controls. Across raters, the overall percentage of correct identifications of suicidal patients was modest, revealing a high rate of both false positives (misidentifying depressed non-suicidal patients as suicidal) and false negatives (failing to identify patients with active suicidal intent).
• Discriminant Validity Limitations: The instrument demonstrated substantial overlap between the verbal protocols of suicidal patients and those of general depressive non-suicidal psychiatric controls. Because a large proportion of the 54 stems explicitly stimulated depressive themes, both cohorts produced responses characterized by dysphoria, exhaustion, and sadness, severely compromising the test’s discriminant validity in isolating self-destructive risk from general psychological distress.
• Confidence Calibration: Efron examined the relationship between raters’ subjective confidence in their diagnostic decisions and their actual objective accuracy. The findings demonstrated a lack of significant positive correlation between judge confidence and diagnostic correctness; clinicians were frequently highly confident in erroneous classifications, illustrating the cognitive biases inherent in unstructured clinical interpretation of projective verbal material.

8. Reliability

In the original 1960 publication, formal psychometric indices of internal consistency (such as Cronbach’s alpha) and temporal stability (test-retest reliability) were not computed or reported. This absence reflected the standard psychometric landscape of the late 1950s, during which projective and semi-projective screening techniques were evaluated primarily through inter-rater agreement and qualitative diagnostic utility rather than classic item-level correlation matrices.

The primary form of reliability relevant to Efron’s evaluation was inter-rater reliability (inter-judge agreement). Clinical psychologists and trainees evaluated identical sets of de-identified protocols. Analysis of the classification decisions revealed low-to-moderate inter-rater consensus. While raters agreed substantially on protocols that contained overt, explicit mentions of death or self-destruction (e.g., direct declarations of wishing to end one’s life), their agreement dropped dramatically when interpreting subtle, latent, or symbolic expressions of despair. Consequently, the absence of an objective, empirically weighted numerical scoring manual contributed to significant inter-judge variance, demonstrating that the test, as implemented in 1960, possessed questionable inter-scorer stability without rigid, objective scoring rubrics.

9. Factor Analysis

No exploratory factor analysis (EFA) or confirmatory factor analysis (CFA) was conducted on the Sentence Completion Test at the time of its construction by Efron in 1960. During that era, computational limitations rendered large-scale factor analysis of open-ended, 54-item qualitative linguistic protocols largely unfeasible in standard clinical research settings.

From a modern psychometric perspective, the absence of factor-analytic validation represents a notable structural limitation. Because the 54 sentence stems were assembled through clinical curation—combining items from antecedent sentence completion batteries with original stems chosen for their face validity in relation to depression—the latent dimensionality of the item pool has never been empirically established. Contemporary structural modeling of similar sentence completion tools indicates that such instruments typically do not conform to a unidimensional construct. If subjected to contemporary CFA, the test would likely reveal a multidimensional structure comprising distinct latent factors: negative self-evaluation, hopeless anticipation of the future, interpersonal alienation, overt aggressive drive, and somatization. Without modern item-response modeling or factor-analytic structural fit indices (e.g., CFI, TLI, RMSEA), the instrument remains an unstandardized, exploratory screening device characterized by high face validity but unverified structural invariance across diverse psychiatric cohorts.

10. Instrument / Measurement Tool

The structural, administrative, and evaluative parameters of the Sentence Completion Test are outlined below based on Efron’s original operational framework:

• Instrument Name: Sentence Completion Test (Efron, 1960)
• Developer: Herman Y. Efron, Ph.D.
• Test Classification: Semi-projective verbal screening device; inpatient clinical screener
• Target Population: Adult psychiatric inpatients (aged 18 and older) admitted to institutional or psychiatric hospital settings
• Administration Modality: Standardized paper-and-pencil; adaptable for individual administration or group testing in institutional wards
• Administration Time: Approximately 30 to 45 minutes, depending on the patient’s cognitive processing speed and psychomotor coordination
• Total Item Count: 54 incomplete sentence stems
• Item Composition:
  • Stems adapted from pre-existing mid-century sentence completion blanks
  • Newly constructed stems explicitly designed to saturate the protocol with depressive, melancholy, and existential themes
  • Neutral, familial, and social stems functioning as baseline comparison stimuli
• Response Format: Open-ended, spontaneous written completion of each linguistic stem
• Scoring and Evaluation Methodology:
  • Qualitative, whole-protocol evaluation by trained clinical raters
  • Categorical sorting into one of three clinical groups: (1) Suicidal, (2) Assaultive, or (3) Non-suicidal/Non-assaultive
  • Rater confidence rating: Judges assign a subjective numerical rating reflecting their diagnostic confidence in the assigned category
  • Absence of an objective, automated linear summative scoring key; evaluation relies entirely on clinical judgment and thematic content analysis

11. Permissions & Fee and Test Year

The Sentence Completion Test was originally published in 1960 in the Journal of Consulting Psychology, an official publication of the American Psychological Association (APA). The formal scientific study introducing the instrument is:

• Publication Year: 1960
• Copyright & Intellectual Property: The scientific paper and associated experimental stems are subject to historical copyright owned by the American Psychological Association. The test was not published as an independent commercial testing kit (such as those distributed by Pearson Clinical or PAR Inc.) and has no commercial fee structure.
• Research & Clinical Permissions: The scale stems as published in academic archives may be utilized for academic, non-commercial, and historical research purposes provided appropriate bibliographic citation is accorded to the original author and journal. Clinical practitioners seeking to utilize sentence completion techniques in contemporary high-risk suicidology settings are advised to utilize standardized, fully validated modern alternatives, as Efron’s battery lacks contemporary normative tables, proprietary manual support, and computerized diagnostic scoring systems.

12. References

• Beck, A. T., Kovacs, M., & Weissman, A. (1979). Assessment of suicidal intention: The Scale for Suicide Ideation. Journal of Consulting and Clinical Psychology, 47(2), 343–352. https://doi.org/10.1037/0022-006X.47.2.343
• Efron, H. Y. (1960). An attempt to employ a sentence completion test for the detection of psychiatric patients with suicidal ideas. Journal of Consulting Psychology, 24(2), 156–160. https://doi.org/10.1037/h0041225
• Freud, S. (1917). Mourning and melancholia. In The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 14, pp. 243–258). Hogarth Press.
• Menninger, K. (1938). Man against himself. Harcourt, Brace & World.
• Rotter, J. B., & Rafferty, J. E. (1950). The Rotter Incomplete Sentences Blank manual: College form. The Psychological Corporation.
• Shneidman, E. S., & Farberow, N. L. (Eds.). (1957). Clues to suicide. McGraw-Hill.

13. Items of the Scale