An integrative attributional perspective of empowerment and learned helplessness: a multimethod field study.

Empowerment has been suggested as a means of facilitating productive and motivated behavior in organizations (Conger & Kanungo, 1988). In general, empowerment is defined as a cognitive state that results in increased intrinsic task motivation (Thomas & Velthouse, 1990). In contrast, Learned Helplessness (LH) has been defined as a debilitating cognitive state in which individuals often possess the requisite skills and abilities to perform their jobs, but exhibit suboptimal performance because they attribute prior failures to causes which they cannot change, even though success is possible in the current environment (Martinko & Gardner, 1982). A comparison of these definitions indicates that they both pertain to cognitive states wherein motivation is either increased or decreased.

Theoretical models of empowerment (Thomas & Velthouse, 1990; Spreitzer, 1995) and LH (Abramson, Seligman, & Teasdale, 1978; Martinko, 1995; Martinko & Gardner, 1982, 1987) have been developed independently, yet they share many similarities, suggesting the possibility that empowerment and LH are polar opposites on a single underlying continuum of motivation.

Although not explicitly stated, the notion that empowerment and LH are reciprocal constructs dates back at least to deCharms' (1968) depiction of a motivational continuum. At one end of the continuum are "Origins" who, like empowered individuals, believe themselves to be in control of their own behavior, and at the other end are "Pawns" who, like LH individuals, believe themselves to be at the mercy of external forces. Similarly, Rotter (1966) distinguished persons with an internal locus of control, who believe that events are contingent upon their own behavior or personal characteristics, from persons with an external locus of control, who believe that events are contingent on forces in the environment not under their own control. More recently, Seligman (1990) contrasted optimists, who view defeat as a challenge to try harder, and pessimists, who view defeat as a permanent setback. His description of optimism and pessimism closely parallels the notions of empowerment and LH. Nonetheless, up to this point, empowerment and LH have been examined separately. The following comparison of independently-developed theories of empowerment and LH notes their striking similarities. This comparison suggests that empowerment and LH may be more appropriately envisioned and more parsimoniously examined as reciprocal constructs.

First, both empowerment (Thomas & Velthouse, 1990) and LH (Martinko & Gardner, 1982, 1987) models make the same ontological assumption (Morgan & Smircich, 1980) that it is necessary to examine individuals' "objective" and "subjective" realities as the bases for understanding and explaining behavior. To illustrate, Thomas and Velthouse (1990) acknowledge the role of objective reality in empowerment, but go on to suggest that individuals' judgments and behavior regarding tasks also are shaped by cognitions that go beyond verifiable reality. In describing LH, Martinko and Gardner (1982) note that cues from the environment are coupled with individual difference variables to influence individuals' interpretations regarding work performance.

Secondly, models of both empowerment and LH explicitly acknowledge their grounding in the S-O-B-C (stimulus, organism, behavior, and consequences) social learning theory approach presented by Davis and Luthans (1980), with Thomas and Velthouse (1990) suggesting that their model resembles the social-learning sequence of stimulus, organism, behavior, and consequences (1990, p. 669) and Martinko and Gardner suggesting that the overall framework for the model is the social learning theory approach (1982, p. 197). It is of note that Spreitzer's (1995) more recent depiction of psychological empowerment continues this approach, containing constructs organized in a sequence of antecedents, psychological empowerment, and consequences. Thus, both empowerment and LH models include similar components, depicting interpretive styles (including attributions) and global assessments from past experiences as influences on affect and behavior.

Third, Thomas and Velthouse note that their model of empowerment is one in which the focus is on intrapersonal cognitive processes (1990, p. 669). Martinko and Gardner use almost the same wording noting that, in their model of organizationally induced helplessness, the intrapersonal cognitive processes of the organism are emphasized (1982, p. 197). Recognizing these similarities between theoretical depictions of empowerment and LH, and because of recent increases in the application of attribution theories to organizational settings (Martinko, 1995), the purpose of this study was to provide an in-depth examination of the dynamics of both empowerment and LH in order to (1) more fully understand the underlying dynamics of each of these separate constructs, and, (2) explore the relationship between these two constructs to determine whether they can be more parsimoniously described as manifestations of the same motivation process. If the second point proves to be valid, the substantial body of evidence that currently exists regarding LH can inform and extend the growing body of literature regarding empowerment. At the same time, the applications-oriented literature on empowerment can facilitate a more pragmatic understanding of implications and theory associated with LH.

Numerous authors have called for multimethod research in organizational studies (cf., Gioia & Pitre, 1990), and there is a trend toward the use of multimethods in other areas of inquiry (cf., Bradshaw-Camball & Murray, 1991). However, in the area of empowerment and LH, data sources have been heavily oriented toward reliance on questionnaires (cf., Spreitzer, 1995), which can suffer from psychometric inadequacies (Schriesheim & Kerr, 1974) and from monomethod bias (Cook & Campbell, 1979). Few studies of empowerment and LH have been conducted in organizational settings (see Mitchell & Wood, 1980 and Spreitzer, 1995 for exceptions), making the generalizability of questionnaires in organizational settings debatable. At this point in time, research on both empowerment and LH in organizational settings appears to be in an exploratory stage. This study addressed these issues by comparing questionnaire data to information from interviews and observations in an effort to combine positivist and naturalist research methods in a field setting. In sum, in this study we explored the dynamics of empowerment and LH and examined whether empowerment and LH might accurately be described as two manifestations of the same construct. We measured empowerment and LH at two points in time using three different measures: questionnaires, semi-structured interviews and subjective researcher evaluations, and analyzed the data with parametric analyses, nonparametric analyses, and two coding schemes. The overall process can be described as a methodological triangulation of both quantitative and qualitative data.

Theory And Hypotheses

Conceptual Issues

As mentioned above, empowerment and LH models contain similar components. These are compared and described in more detail below and are integrated into a single model depicting the relationships in Figure 1.

Interpretive styles. Recent cognitive models depict both empowerment (Thomas & Velthouse, 1990) and LH (Martinko & Gardner, 1982, 1987) as the individual's subjective realities based on intrapersonal interpretations of events. Both models include attributions as a key part of the individual's interpretations. While there are many specific attributions that can be made about events, Abramson, Seligman, and Teasdale's (1978) reformulation of learned helplessness theory presents four dimensions along which attributions can be characterized: locus of causality, stability, globality, and controllability. Locus of causality refers to the location of the cause of an event: internal or external to the self. Stability refers to the expected duration of the cause, with a stable attribution representing a cause of long duration and an unstable attribution representing a cause of short duration. Globality refers to the pervasiveness of the cause in other situations, with a global attribution representing a far-reaching cause, and a specific attribution representing a narrow-ranging cause. Controllability refers to whether the cause of an event is attributed to be controllable by the actor or uncontrollable. Controllability is an important dimension, as Greenberger and Strasser (1986) propose that LH individuals no longer believe that they can control the outcomes they experience. Since Abramson et al. (1978) presented these attributional dimensions, they have subsequently been used in a number of studies of attributions and these same dimensions are included in models of empowerment (Thomas & Velthouse, 1990) and LH (Martinko & Gardner, 1982). In fact, attributions occupy such a central role in depictions of LH that they have been used as surrogate indicators of LH (Peterson, Semmel, vonBaeyer, Abramson, Metalsky, & Seligman, 1982; Kent & Martinko, 1995).

Thomas and Velthouse (1990) are not specific about the nature of the attributions associated with what they refer to as an individual's interpretive processing of events (1990, p. 671). However, Seligman (1990) suggests that optimists believe that bad events are temporary (unstable), have specific causes (specific), and that those causes are external. The relationship of these attributions to empowerment has not been examined previously; however, habitual styles of making attributions have been shown to be predictive of LH. Specifically, consistently making internal, stable, global and uncontrollable attributions for negative situations is related to LH (Seligman, 1990). Thus, attributions appear to be a key explanatory variable in the depiction of LH, and both Seligman's (1990) and Martinko's (1995) descriptions of the attribution styles of optimists suggests to us that the attribution styles of empowered individuals are the reciprocal of the styles of LH and pessimistic individuals.

Affect. A number of different emotions have been found to be associated with LH (Peterson & Seligman, 1984), most notably depression (Coyne & Gotlib, 1983; Seligman, 1990). Diener and Dweck (1980) found that LH individuals quickly begin to express negative affect when faced with failure. While the affect associated with LH in organizations has not yet been specifically investigated, the affect theorized to be associated with LH in general is depression, anxiety, stress, frustration, hostility, fatigue, anger, shame, and alienation (Martinko & Gardner, 1982, 1987; Peterson & Seligman, 1984).

If it is indeed the case that empowerment and LH are reciprocal expressions of the same construct, then we might anticipate that empowered persons would experience affect that is the reciprocal of the affective states described above. That is, whereas LH persons are expected to be depressed, anxious, stressed, and frustrated; empowered persons would be expected to be happy, calm, stress-free, energized, and connected. However, Seligman's suggestion that, "if your explanatory style is optimistic, your depression will be halted" (1990, p. 75) seems to imply that the reciprocal of the negative affective states associated with LH may not necessarily be the opposite affective state. That is, when dealing with negative events, the reciprocal of depression may not be happiness, but may simply be an absence of depression. Similarly, the reciprocal of frustration and anger may be a lack of frustration and anger. We …