Self-esteem is generally considered to be how individuals evaluate their self-worth and competence. This evaluation can be positive or negative. Having a positive sense of self or good self-esteem is linked to many positive behaviors, such as achievement, initiation, motivation, and good mental health. Not all researchers and clinicians, however, approach the study of the self or self-esteem in the same way, and this lack of consensus has led to the many measures that are available for trying to ascertain the level of self-esteem of individuals.
In the next sections of this entry the various ways of thinking about the self and self-esteem will be reviewed. There will be a brief history of self-esteem and a discussion of how to define self-esteem (and related terms). Finally, the issue of how to measure self-esteem and various measures that are currently being used to study self-esteem will be reviewed.
A Historical Overview of Self-Esteem
The first glimpses of theoretical differences in explaining the self system can be seen beginning with the mind-body debates of the Greek philosophers, Socrates, Plato, and Aristotle. These philosophers all theorized about where the “essence of a person” (the self) resided, in the soul or in the mind. Centuries later these early quests to understand the body and soul influenced other philosophers and scientists to examine what constitutes an individual’s “self-identity.” One of the most influential of these philosopher-scientists to discuss the self-concept in depth was William James in his 1890 book The Principles of Psychology. James is credited as the first psychologist to develop a theory of self-concept. It is his theoretical structure of self-concept that laid the groundwork for all of the other theories that have since been posited about this construct.
James incorporated his ideas of self into the concepts of “I” and “me.” The “I” is the subjective self and the “me” is the objective self. The “I” is the essence of the self or what constitutes one’s personal identity. It cannot be observed directly and, according to James, can only be discovered through reflection. The “me” is the combination of all things that are objectively known about the self; that is, those things that incorporate one’s material self, family, and friends and one’s social self. Unlike the “I,” the “me” can be observed and empirically examined.
There are three elements of the Jamesian “me”: the material “me,” the spiritual “me,” and the social “me.” According to James’s theory, these elements are ordered in a hierarchical structure with the material self at the bottom, the spiritual self at the top, and a combination of material and social elements in the middle. Self-evaluations then, according to Jamesian theory, are the reflection of how individuals weight the subjective importance of these different characteristics of the “me.” The importance placed on these characteristics becomes the standard that objective information is referenced against and dictates what is considered to be a success or a failure.
This overview of James’s theory regarding self-concept is only a brief summary of a very critical piece of work that has affected self-concept research throughout history, including contemporary times. Indeed, symbolic interactionists Charles Cooley and George Herbert Mead were influenced by the Jamesian concept of the “social self” when developing their theories on the impact of socialization. Cooley’s concept of the “looking-glass self,” for example, posits that individuals can only know themselves through the reactions and social interactions of others. Mead also considered social interactions important to the development of the self but, like James, acknowledged a core self that is modified by these interactions. Other psychological paradigms as diverse as psychoanalytic and behaviorist theory were affected by James’s framework of self-concept. Sigmund Freud and his followers, like James and Mead, believed that individuals were influenced by the way others view them or feel about them. They, however, added the idea that the self-concept exists at birth and is modified by childhood interaction with family members, especially the mother.
In contrast to the psychoanalytic theorists, behaviorists do not believe that individuals have an innate self. Instead, behaviorists such as John Watson and B. F. Skinner believed that individuals are the sum of all of their experiences, which are modified by positive and negative reinforcement from the environment. Individuals, therefore, interact with their environment and, through various forms of reinforcement, learn what is beneficial to their survival and continue those behaviors. By understanding these behaviors and social contingencies, individuals can begin to understand how they are different from others, and derive a sense of self. Therefore, self-development, according to behaviorism, occurs as individuals become active in creating an environment that is positively reinforcing to them.
These theoretical paradigms were not the only ones affected by the Jamesian structure of the self. Over the course of the last 3 decades, the research has moved away from the view that learning is affected by the environment only. Learning is now viewed as an interaction of cognitive processes and reinforcement from the environment. The area of self-concept research has followed the lead of the field. It is clear that the concept of self is part of the cognitive structure of a person. When individuals think of themselves, they have a mental representation of themselves. When they interact with others, they derive information from these encounters and modify how they view themselves and how they view others. In this way, individuals are processing information from their environment and either modifying existing thoughts or ideas or adding new ones. This information processing during important interactions is what is considered social cognition. Most of the theories from the literature today are based on some concept of social cognition.
The problem that social cognition addresses is whether individuals are accessing information about the self when answering questions or responding to statements that seek to gather information about either self-concept or self-esteem. Thus, when attempting to gain any information from measures where individuals report on their own thoughts and feelings (self-report), the researcher assumes that some aspect of memory is being incorporated in the response. In general it is believed that information about this self is held in long-term memory. Some researchers, however, believe that the only way to answer questions related to self is to have this information assessable in short-term memory. Cognitive researchers make the argument that information on emotions, plans, and evaluations are generally held in short-term memory and, therefore, should be accessible during self-report assessments. They base their research on a general information-processing model that emphasizes the retrieval of information from the short-term memory.
Self-Esteem Definitional Issues
Self-definitions can be classified into uni- and multidimensional definitions. Early definitions of self-concept emphasized a general or overall self-concept that influenced individuals’ reactions or behaviors over all situations. For example, people with a high sense of self-worth will deal more competently with all life events than will those who have a low sense of self-worth.
Since the 1980s, however, there has been a general shift away from these unidimensional definitions toward a more multidimensional view. The multidimensional definitions emphasize that individuals view themselves differently across situations. One may have a good self-concept of his or her reading abilities but not have one involving his or her math abilities. Hence, the behavior would be different depending on whether a person was dealing with a reading or math situation.
The dimensionality distinction is not the only one that has affected research on the self. A more difficult distinction to make when defining the self is whether it is evaluative or descriptive. Some researchers see this distinction as the difference between self-esteem and self-concept. Self-esteem, for example, is viewed as the degree of liking or satisfaction with the self (e.g., “I am a good person”). Self-concept, on the other hand, is more descriptive and can be evaluative (as in “I love math”) or nonevaluative (as in “I am a boy”). Unfortunately, the difference between these two constructs is often just semantic. Individuals can describe themselves as both descriptive and evaluative at the same time (e.g., “I am smart because I like to read”). It is important when defining the self to be explicit about which element of the self is being studied, the descriptive or the evaluative. Too many researchers are not making these distinctions and, therefore, are claiming to study the self-concept or descriptive self when they are actually gathering information on the evaluative self. Some of the confusion in these terms comes from the historical study of the self where philosophers and researchers tried to understand the nature of the self and its evaluative component, self-esteem.
Self-Esteem Methodological Issues
Classically, the way to measure the self and self-esteem is through some type of self-report methodology. Generally, self-report methodologies can be categorized as either reactive or spontaneous. The most often used as well as the more popular of these two methodologies is the reactive self-report. This type of self-report is characterized by a closed-ended response format that uses a Likert-type scale representing differing dimensions. In general, participants in a study are asked to react to questions or dimensions derived by the researcher and then to locate themselves on the scale provided. This type of methodology is usually easy to use, often creates little interaction between the participants and the researcher, and creates data that can easily be manipulated quantitatively. The majority of self-concept measures used in current research are considered reactive.
The spontaneous approach allows the participant rather than the researcher to create the dimensions that are examined. This approached is often labeled “open-ended” because participants are allowed to answer in unconstrained ways. In this approach, participants generally answer a vague question such as “Who are you?” without any constraints as to how long they answer or how much they answer. In some situations boundaries are established for the amount of time or numbers of answers that are given. The spontaneous approach, generally, is more difficult to score and summarize than is the reactive and usually involves a great deal of interaction between the participant and the researcher.
The decision of what type of methodology to use is dependent on what type of process the researcher is attempting to examine, descriptive or evaluative. Some researchers believe that these two processes tap into different aspects of the self. Hence, by using only evaluative measures, important information about the construct of the self is lost. Thus, it is important to understand which aspect of the self is being studied, and this is an important distinction to keep in mind when deciding on what type of methodology to use when examining self. Other issues that are also important to the measurement decision are discussed below.
Even though self-report is not the only methodology used for gathering information on children and adults, it is often the most used in the literature and consistent with theoretical concepts on how to retrieve information about the self. There are inherent problems, however, in dealing with this type of methodology. Issues such as how accessible this information is in the memory when being asked these questions, the situations or contexts a person is in when answering these questions, and other individual or developmental differences that might influence how one answers questions about the self are important factors to consider.
The accessibility of the self refers to the cognitive processes involved in how information about the self is distributed and processed through the memory and how easily it is retrieved when queries regarding the self are made. Due to the reliance on memory in examining the self, there are inherent biases such as self-deception that occur when trying to access this information.
In general, because obtaining information about the self involves cognitive processes that have to be accessed and organized, researchers who study the self need to be aware of the biases that can occur in self-reports. In general, information that is easily accessible, that involves information about the self that is attended to on a regular basis, and that agrees with standard norms is more likely to be reported than information that does not meet these criteria.
The second area of concern in using self-reports involves the contextual, situational, and cultural factors that influence how individuals respond to questions about the self. If individuals are asked questions about the self when they are at work then they are more likely to discuss themselves in terms that define them in that environment. Information about the self can also be affected by the emotional state (good or bad moods) of an individual, how individuals want to be perceived by the researchers (self-presentation), and the culture of the individuals. The view that context, situation, and cultural factors influence the self is highly consistent with the symbolic interactionist ideas that were posited by Cooley and Mead. When studying the self, it is important to understand the environmental and cultural norms that may be affecting the responses that individuals give to questions regarding the self. These factors may be contributing to the actual development of self-esteem or may be biasing the responses that are given.
Almost everyone will be given some form of a self-esteem measure at some point in their life. Self-esteem plays a central role in our understanding of human behavior, and it is important that care is taken in choosing the right instrument for measuring this characteristic. The following section examines a representative sample of self-esteem measures that are widely used and readily available.
Perhaps the most famous and widely used unidimensional measure is the Rosenberg Self-Esteem Scale that was developed to study adolescents but is now used across the whole age range. The Rosenberg is a 10-item scale that asks individuals to rate themselves using four categories (strongly agree to strongly disagree) on a series of questions regarding their sense of self-worth or confidence (e.g., “I feel that I’m a person of worth, at least on an equal plane with others”). This scale has been widely used in all areas of psychology, is easy to administer, and can be adapted to be used with children as well as adults. It is considered a highly reliable and valid measure of self-esteem. The measure is especially useful for getting a global sense of how a person feels about him- or herself and is a good predictor of other measures of mental health such as depression and anxiety. It is not a good predictor of more specific domains like academic ability or body image.
Another widely used unidimensional measure is the Coopersmith Self-Esteem Inventory. This inventory was developed for use with children but now is used with both children and adults. The inventory assesses subjects’ general attitudes toward themselves and in comparison to significant others in their lives. This scale contains around 50 items (even though this has been adapted across time) and individuals rate these items as being like me or not like me. The Coopersmith has good reliability and validity. Like the Rosenberg, this scale has been used effectively to ascertain a global sense of self-worth but not in predicting more specific domains of self-evaluations.
Overall, unidimensional scales are appropriate and useful for gathering information on the general mental health of individuals. These scales were developed and based on assumptions that global self-esteem can be predicted across domains and be influential even to specific evaluations of domains such as achievement. However, psychological research has found that this is not the case, and a more multidimensional scale needs to be used when examining specific domains of functioning (e.g., achievement or sports).
As with the unidimensional scales, the multidimensional scales that are mostly widely used originated with work with children and adolescents and were adapted for use with other older and younger populations. These scales assume that individuals evaluate themselves differently across important areas of their lives. For example, they may evaluate themselves negatively on academics but positively on physical activities. The most noted of these scales is the Harter Self-Perception Profiles, which were originally designed for children but have been adapted and validated for adolescents, college students, adults, the elderly, and learning disabled students. In general, this measure assesses domains of cognitive, social, and physical perceived competence as well as general feelings of self-worth. The domains measured and numbers of items vary by the form that is used, but in general the scale contains 28 items. Each item consists of two paired alternative statements regarding the domains that are being measured (e.g., “I am good at reading” or “I am not good at reading”). Individuals are instructed to pick one of these items and rate it using response categories range from really true for me or sort of true for me. These measures have good reliability and have been validated across multiple studies.
The Self-Description Questionnaires created by Herbert Marsh and colleagues is similar to the Harter scales in that multiple dimensions are examined and the scales have been found to be reliable and valid for use with individuals ranging from childhood into adulthood. The dimensions vary across the age range, but in general scores can be obtained on individuals’ self-evaluation of academic ability, physical ability, physical appearance, relations with same and opposite sex peers, relations with parents, religion/spirituality, honesty, emotional stability, and general self-worth. Depending on the age of the respondent, the measure varies from 64 to 134 items regarding ability (e.g., “I am good at running”) that are rated from as definitely false to definitely true. The scaling varies from 5-point scales for the youngest children to 8-point scales for adolescents and adults.
These scales are just a small example of what is available for examining self-esteem. They were chosen because they have been used across multiple populations, and due to their popularity, there is information on how reliable and valid these measures are for use with various groups in the population. However, a simple search of the literature in self-esteem will uncover many more scales that are available for examining various aspects of self-esteem. Care should be taken in ascertaining that these measures are appropriate for use with various populations and have been shown to measure what they purport to measure. It is important for researchers and clinicians to use measures that are specific to the type of issue they want to understand, whether it be a global sense of self-worth or more specific evaluations of competencies across a wide spectrum of behaviors.
Over the course of time the theoretical paradigms involving the self have taken their structure from the early works of William James and have built upon each other to form the contemporary theories that today are based on cognitive processing of aspects of the self. However, debate about what constitutes the self continues. Historically, the idea that social interactions play an important role in the development of the self has had strong support, but the exact nature of what constitutes the self remains vague. What is generally agreed upon, however, is that understanding the self involves understanding the separation of the descriptive and evaluative self. This distinction has a direct impact on the methodology that is used to gather information on the self.
The methodological issues are an especially difficult problem for research on the self. Even though the research history on self and the self-concept is vast, there are still many such issues on how to best measure this phenomenon. It is clear that the self develops across time and also changes (perhaps frequently) during that time, and thus measures that can examine this development are important but are difficult to find and confounded by the abilities of each age group. Many of these issues can be resolved by paying strict attention to the research methodology and psychometrics involved in obtaining information from self-report methodologies. Others, however, are more complicated and involve the theory and definitional differences in those who study the self and related constructs.
The problems of measuring a construct like self-esteem are inherent in all studies that try and examine and predict elements of the personality. Some researchers assume that phenomena that occur at the cognitive level will be manifested by behaviors that can be observed and measured. Others feel that behaviors should not be used to measure personality traits because we are unable to determine what cognitive processes might be affecting that trait. They argue that constructs like self-esteem should not be studied because there is no clear way to measure them or even know if you are measuring the right thing. Most people, however, can name many situations where they have observed behaviors of children, friends, and family members and suspected that these behaviors emanate from some activity involving their self-esteem. Therefore, there are observable behaviors that can be directly attributable to how people conceive of themselves.
As the majority of the theories suggest, individuals develop their self and self-concept through social interaction. There may be some innate cognitive process that is modified by social interaction or just social interaction, but either way people display behaviors and express opinions about who they are, from a very young age. For this reason it is important that psychologists be able to study and measure this phenomenon. Future work needs to address the issues of theory, definition, and methodology that have remained unresolved across the centuries. A consensus on these issues will lead to better measurement and a better understanding of how self-esteem influences the mental health and behavior of individuals.
- Baumeister, R. F. (Ed.). (1993). Self-esteem: The puzzle of low self-regard. New York: Plenum Press.
- Blascovich, J., & Tomaka, J. (1991). Measures of self-esteem. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes (Vol. 1). San Diego, CA: Academic Press.
- Davis-Kean, P. E., & Sandler, H. M. (2001). A meta-analysis for preschool self-concept measures: A framework for future measures. Child Development, 72(3), 887-906.
- Harter, S. (1983). Developmental perspectives on the self-system. In P. H. Musser (Ed.), Handbook of child psychology (4th ed., Vol. 4, pp. 275-385). New York: Wiley.
- Hattie, J. (1992). Self-concept. Hillsdale, NJ: Lawrence Erlbaum.
- James, W. (1950). The principles of psychology (2nd ed., Vols. 1 & 2). New York: Dover. (Original work published 1890)
- Wylie, R. C. (1989). Measures of self-concept. Lincoln: University of Nebraska Press.