|
|
| (26 intermediate revisions by the same user not shown) |
| Line 1: |
Line 1: |
| =Introduction= | | =Introduction= |
| [[File:Three models.png|right|250px]]For decades, researchers and the personality assessment industry have studied and refined various methods for measuring personality in organizations. These methods are used for hiring, personal development, coaching, and more, including in clinical settings. We at GRI have reviewed these methods by analyzing their effects and benefits on individual and organizational performance, how they were developed, and by studying their users: traditional users like human resources managers, recruiters, coaches, and clinicians, as well as managers and executives. | | [[File:Three models.png|right|250px]]For decades, researchers and the personality assessment industry have studied and refined various For decades, researchers and the personality assessment industry have studied and refined various methods for measuring personality in organizations. These methods are used for hiring, personal development, coaching, and more, including in clinical settings. We at GRI have reviewed these methods by analyzing their effects and benefits on individual and organizational performance, how they were developed, and by studying their users: traditional users like human resources managers, recruiters, coaches, and clinicians, as well as managers and executives. |
|
| |
|
| In social matters, unlike in technology, change happens slowly. As for assessment techniques, particularly with personality, as their history shows, understanding their true potential takes time<ref>See [[Personality_Assessment_Story_and_Prospect|here a brief history, and some perspectives about the progress of personality assessments.]]</ref>. This extended process may be due to a combination of factors: the abstract nature of the concepts being measured, the effort needed to learn these techniques, and the emotional attachment people develop toward them. | | In social matters, unlike in technology, change happens slowly. As for assessment techniques, particularly with personality, as their history shows, understanding their true potential takes time<ref>See [[Personality_Assessment_Story_and_Prospect|here a brief history, and some perspectives about the progress of personality assessments.]]</ref>. This extended process may be due to a combination of factors: the abstract nature of the concepts being measured, the effort needed to learn these techniques, and the emotional attachment people develop toward them. |
| Line 9: |
Line 9: |
|
| |
|
| =The Three Models= | | =The Three Models= |
| Three models clearly emerge from analyzing assessment techniques used in organizations and research labs that account for the majority of them. The three models are the following: | | Three categories of models clearly emerge from analyzing the assessment techniques used in organizations and research labs, taking into account their characteristics and uses. 28 criteria have allowed us to differentiate assessment techniques that are particularly applicable to social behavior<ref> See [[Assessment_Techniques_-_Characteristics | here details about the 28 criteria, their definition, and regroupement in four grand categories.]]</ref>. The three categories appear as the biggest differentiators from both conceptual and practical standpoints. |
| * '''Factor models'''. The factor approach emerged in the 1950s but did not gain widespread acceptance until research from the 1970s to the 1990s promoted a limited number of factors to explain behavioral traits. We had to wait for the 2010s to start seeing their use in organizations.
| |
| * '''Trait models'''. These models were among the first to be used in psychiatry and clinical practice in the late 1800s. They were soon employed for large-scale recruitments in the 1920s, and later for executive search.
| |
| * '''Type models'''. Although the first typologies date back to ancient times, their statistical evaluation is more recent, starting in the 1940s, and their use has grown significantly with the rise of coaching in the late 1980s.
| |
|
| |
|
| The three models offer different understandings of how people function from behavioral, emotional, and cognitive standpoints. Although each can be useful for addressing questions about people, they also have limitations and should be used wisely. Seeing people in context, or not, and with varying nuances, the three models ultimately carry a different vision of how people can be understood, how they possibly show up, grow, and perform. | | The three models presented below offer different understandings of how people function from behavioral, emotional, and cognitive standpoints. Although each can be useful for addressing questions about people, they also have limitations. Seeing people in context, or not, and with varying nuances, the three models ultimately carry a different vision of how people can be understood, how they possibly show up, grow, and perform. |
|
| |
|
| ==Measurement and Representation==
| | The analogy of a two-dimensional map is often used to illustrate the importance of accurately understanding or locating behaviors. We also use it to explain the differences among the three models, to understand the potential challenges and efforts required to move from point A to B. Above all, those models speak to action and, at least partially, account for how action unfolds as we navigate the map. |
| The three models, Factor, Trait, and Type, are used to both measure and represent behaviors, but the two aspects, measurement and representation, need to be considered separately, because a technique may use one model for measurement, but its representation may be based on one or a combination of the other two models. | | The three models are the following. Each is detailed in a separate article. They are briefly presented below. |
|
| |
| For instance, some techniques measure factors and are thus factor-based. Whatever scales they use, which may vary greatly between the decile and standard deviation scales, they convert the measures of factors into traits or types and use the results as such, rather than using the factors themselves. The same happens with techniques that measure traits, which are represented as types.
| |
|
| |
|
| The techniques end up being used as they are represented, but not for what they measure. Because of the combinations we see among the models, we differentiate them based on how they are both measured and represented.
| | {| class="wikitable" style="margin: auto;" |
| | | ! Model !! Description |
| ==Adaptation==
| | |- |
| Adaptation is an important aspect of what the three models deal with, because the object of their measure and representation, behavior, is in constant flux. There is always something happening out there: a critical or minor event, radical change, disruption, crisis, drama, etc., and also from inside: health issues, lack of sleep, just a bad day, personal challenges, etc., that constantly ask us to act, as actors do on stage.
| | | [[Factor_Model | Factor]] || Behaviors are precisely measured by their coordinates on the map. The distance and effort required to cover distances on the map can be anticipated. |
|
| | |- |
| Additionally, there are always expectations, from ourselves and others, including management or HR, or from our role in the organization at large, to behave in a certain way and adapt our behavior accordingly.
| | | [[Trait_Model | Trait]] || Behaviors are estimated by recouping small areas on the map. One knows we are more or less distant from it. Distances to be covered can only be roughly approximated. |
|
| | |- |
| This important aspect of adaptation has repercussions for how one envisions a person’s growth, provides understanding and support, and ultimately, how the person takes on an assignment or a job and performs, including on an emotional level. Development happens differently for everyone, an aspect revealed with varying degrees of accuracy by the three models. This aspect is discussed below for each.
| | | [[Type_Model | Type]] || Being regrouped in broad categories, behaviors on the map appear as large areas, making it impossible to predict the distances to cover. |
| | |} |
|
| |
|
| =Factor Model= | | =Factor Model= |
| A factor model helps position a phenomenon along continuums represented by lines extending from one end to the other in two opposite directions. It helps answer questions about how the phenomenon will develop along the continuums. For instance, with localisation of a person on a map: how does one go from A to B or away from B. Or temperature: how does temperature increase or decrease? With skills: how can this person develop or lose that skill?
| | [[File:Factor_San Francisco.png|right|200px]] |
|
| | With the factor model, as the illustration on the right shows, behaviors are precisely measured, like knowing latitude and longitude on a map. When moving from one point to another on the map, it is possible to anticipate the route, calculate the distance, and select the best way. |
| In the behavioral domain, the factor model addresses questions about how people behave, how they adapt, and the resources required to do so. When adequately designed, it applies at multiple levels, not only the individual one but also the job, team, company, and industry levels. This helps understand how individuals relate to their environment, how they adapt to it, and, vice versa, how environments relate and adapt to people.
| |
|
| |
|
| ==Representation==
| | The factor approach emerged in the 1950s but did not gain acceptance until research from the 1970s to the 1990s promoted a limited number of factors to explain behavioral traits. Systems based on factors work with a limited number of them. We believe at GRI that four is the optimal number. Some models in research work with two, three, five, and up to seven factors. |
| [[File:Factor_San Francisco.png|right|250px]]
| |
| A factor model can best be represented on a two-dimensional map, where an object or person can be located using two axes. As shown on the right, the latitude and longitude will help locate the red dot at the center of Union Square in San Francisco. A third factor, altitude, would help locate the object in other maps in space. Another factor, time, would add a fourth dimension to their location.
| |
|
| |
|
| As with geolocation, social behaviors may be characterized by a limited set of factors. Using factor analysis, words and expressions depicting social behavior can be grouped into four clusters. Each cluster contains a latent hidden factor that can be represented by a continuum, like the longitude and latitude.
| | When used to understand behavior adaptation and individual growth, the factor approach identifies where development starts, where it ends, how it happens, and the effort required to sustain the adaptation. |
|
| |
|
| A person’s factors can then be located on four continuums, using a standard deviation scale, as shown in the illustration below<ref>These paragraphs are a very condensed introduction of personality factors as they emerged over several decades of research, and have proven to be the most effective way to measure and represent social behavior.</ref>. Like latitude and longitude, which can locate an object on a map, the four factors add nuance to one another to describe a person’s social behavior. An individual’s specific behavior will be located by combining the four factors and analyzing them together.
| | Potential users for the factor model are executives, managers, and other users of types and traits (see below). Applications of the factor model in organizations include management, leadership, and organizational development, as well as uses related to types and traits (see below). |
| [[File:Four Factors Combined.png|center|400px]]
| |
|
| |
| A unique profile can be created by connecting the four factors with lines, as shown below. This presentation of the four factors side by side helps illustrate their relative intensities and mutual influence. The visual characteristic of the profile, its shape, instantly combines the four factors, and, once learned, can gather the full meaning of the combinations. One can then use the profiles and continue learning from them and the people they represent.
| |
| [[File:Social Behavior Profile.png|center|400px]]
| |
| | |
| The relevance of the profiles spans across various individual and organizational levels. As we do with the adaptive profiles at GRI, three profiles like the above combine, one representing a person’s natural behavior that is expressed most spontaneously, the adaptation of the person to the environment, and how the two translate into the effective behaviors: the behavior most probably observed by others.
| |
|
| |
| The profiles can be used to represent the behaviors expected in a job, from a team, and from a company, or to describe behaviors at an industry and societal level. At all these levels, the words and expressions used to describe behaviors can be regrouped into the same clusters, as for people, and represented using the same four factors. When designed this way, the measures and representations from the assessment technique help analyze differences in social behavior across levels, to understand the efforts required for adaptation, and to envision how the adaptation will unfold.
| |
| | |
| The above representation of the four factors together has been adopted, with minor variations, by several techniques since the 1950s. Some evolved with the representation of types and traits, keeping the factor model under the hood. Others, like GRI, both measure and use factor representations, bringing the power and precision of the profiles closer to strategic decision-making.
| |
| | |
| ==Measure==
| |
| A factor is a measure on a continuum represented by a horizontal line and two poles, or more precisely, vectors that run in opposite directions, which can locate various expressions of a phenomenon that are close, distant, at opposite ends to each other, and at different levels of intensity. The two extremities are opposite and extreme expressions of the phenomenon, with a neutral value between them, represented by a triangle.
| |
| | |
| For instance, how hot or cold it is can be represented on a continuum, as illustrated below. The following adjectives: glacial, freezing, chilly, cool, lukewarm, tepid, torrid, steaming, roasting, boiling, etc, could locate the temperature on the continuum at different levels of intensity.
| |
| [[File:Factor_temperature_scale_dot.png|center|500px]]
| |
| | |
| When estimating room temperature, the neutral value may be close to the body temperature, while cold and hot may be moderately, very, or extremely distant from it. The red dot in the illustration indicates where a temperature measurement or best estimate may be.
| |
|
| |
| The scale used by an instrument may be similar to the one above, which duplicates the standard deviation scale we use at GRI for behaviors. Interval scales are also often used in personality assessments to measure behavioral intensity.
| |
| | |
| As discussed below, most individual and group characteristics used in organizations, such as competencies, skills, abilities, and experience, have generally been represented by traits and types. However, social behaviors and emotions, like finding a location on a map, can be more precisely defined in terms of their factors.
| |
| | |
| ==Factors’ Measurement and Use==
| |
| Factors have typically been measured using adjective lists, whose analysis was at the origin of their discovery, and through open scenarios in which participants decide whether certain adjectives are part of their life experiences. Other techniques use questionnaires with forced-choice scenarios, a technique more common in trait assessments.
| |
|
| |
| In any case, these techniques enable the positioning of the four factors of an individual’s behavior along each of their continua. They have also allowed for measuring and representing an individual's behavior in their past, current, and immediate contexts, as well as how they engage within their environment.
| |
|
| |
| The use of the factor model includes trait and type models in recruitment, coaching, and team building (see below). It helps to take the analysis to jobs, refining the behaviors expected in them, and bringing greater rigor to understanding how a person will fit the job or how the job needs to adjust to them. The factor model offers greater sophistication and precision in representing behaviors than the trait and type models, enabling its use in organizational development by combining and analyzing multiple profiles across different levels of the individual, job, team, company, and industry.
| |
| | |
| The factor model is the one among the three that offers the greatest capacity for developing people. As discussed below, our brain and senses generally first use the type and trait models. Those two easily preempt what might otherwise come from a factor model, until it’s learned and used.
| |
| | |
| ==Adaptation Efforts==
| |
| On a 2-dimensional map, the factor model can indicate where one is situated and where one needs to go, as well as the effort required to get from one location to the other.
| |
| | |
| Similarly, for behaviors, when going from A to B, the four factors together, within the profile, once measured, will inform what the behaviors are at the beginning and the end. Since the measures are behavioral, they will also indicate how this will happen.
| |
| | |
| By comparison, the trait model can only indicate how close one is to the target location, but only imprecisely in its surroundings. And the type model only indicates whether one is within the target location's region. Additionally, the two models cannot determine the effort required to go from A to B, nor how that journey will unfold.
| |
|
| |
|
| =Trait Model= | | =Trait Model= |
| The trait model is the most widely used in recruitment applications and clinical psychology. A trait can be any individual characteristic that you may think about, labeled with one word or a short expression.
| | [[File:Trait_San Francisco.png|right|200px]] |
| | | As illustrated by the red circle on the right, measures from a trait system indicate how far one is from a target, but that can be from different locations that we don’t know. Additionally, the route and time between locations can only be a rough but impractical estimate; we cannot know either. |
| Traits help answer questions about the attainment or proximity to a phenomenon or a concept, such as with hot temperature: How hot is it out there? Or with skills: How much skilled is this person? Or with creativity: How much creative is this person?
| |
|
| |
| By doing so, trait measures focus on a phenomenon of interest, disregard the occurrences or expressions that are distant, and which, compared to the factor model, are eventually on the other side of the continuum.
| |
| | |
| Traits are easy to understand and use because they eventually correspond to how questions are naturally asked and answered. Traits are omnipresent. Except for clinical applications, traits usually require little explanation, can be learned fast, or may only require a reminder of their definition from the Internet.
| |
| | |
| ==Representation==
| |
| [[File:Trait_San Francisco.png|right|250px]] | |
| With most techniques, the further the measurement is from the targeted trait, the more different it may be, in many ways. This is shown in the map representation on the right, where the red rim indicates the distance to the targeted behavior. The trait’s measure may be anywhere on the rim.
| |
|
| |
|
| The intensity of a trait is typically represented on a decile scale. The maximum value of 10 indicates that the trait most probably and intensely applies to the person. The null value indicates that it doesn’t apply at all. The more the person is characterized by the trait, the more intense and the higher the measure is on the scale. | | The trait model was the first to be used in psychiatry and clinical practice in the late 1800s. It was soon employed for large-scale recruitments in the 1920s, and later for executive search. The number of traits that a system from this model can measure is unlimited. It can be as low as one trait in clinical applications. In organizational applications, one trait can be leadership, creativity, or honesty. In recruitment, the number may be as high as 30. This number can easily double or triple with batteries of tests. |
|
| |
|
| As in the example below on the right, 4.7 indicates the trait’s intensity on the decile scale. The illustration on the left more clearly shows, in a flat representation, that values away from the target are not at the opposite end of a continuum. They may be when the techniques actually measure factors and represent them as traits. This aspect is sometimes represented as a cone in a three-dimensional space.
| | Trait models are not built to understand a person’s development but measure the starting and end points of a development, not how it will occur. That’s another major difference with the factor model. |
| [[File:Factor_Circular and Bar.png|center|300px]]
| |
| | | |
| As shown below, traits measuring social behaviors often come in groups, represented as a bar chart(left) or a spider chart (right). The higher the bar, the more intense the trait is. The two charts give a different impression of the traits' proximity. Traits that are close and on the same side of the spider chart appear closer than those on the other side. This is also how systems that use this representation organize traits around a wheel.
| | Potential users of trait systems include clinicians, recruiters, and coaches, but not executives and managers directly. Trait systems are used in organizations for recruitment and vocational guidance. They are also used like type systems (see below), but rarely cover the applications seen with the factor model (see above). |
| [[File:Trait_basic_multiple spider.png|center|700px]]
| |
| | |
| The charts suggest that traits are distinct, but treating them as factors reveals that they overlap and should be regrouped. The tendency has thus been to keep the number of traits to a minimum, close to four or five, as it emerged from factor model studies.
| |
| | |
| Everyone develops a sense of what most traits mean. But a trait’s measure will be defined and analyzed differently by different people. Additionally, a concept can manifest in various ways, a phenomenon that cannot be captured by a single trait.
| |
| | |
| ==Measure==
| |
| [[File:Trait_Hot not Hot.png|right|100px]]
| |
| A trait is measured on a continuum represented by a line with one pole, or a vector generally pointing upward, that locates how close the measure is to the phenomenon of interest as conceptualized by a person or a system.
| |
|
| |
| The phenomenon of interest is at the top of the vector; its absence is at the bottom, with a null value, suggesting that the phenomenon may not occur at all or in a way that’s totally off focus.
| |
|
| |
| In the example on the right, the focus is on “how hot is it?” which is represented at the top of the vector. The bottom values, which may indicate whether it’s cold, freezing, or just tepid, are irrelevant to answering the question.
| |
|
| |
| The trait model, with its continuum and bottom value, applies to many physical phenomena, such as mass, force, weight, length, speed, acceleration, mechanical power, and electrical power. It also applies to many concepts in social sciences, such as skills, competencies, abilities, experience, interests, and even cognitive abilities.
| |
| | |
| ==Traits’ Measurement and Use==
| |
| Traits are used in recruitment, coaching, personal development, and to characterize what’s expected in jobs. They are also used for clinical assessment. For instance, the following are traits used in recruitment and coaching: Agreeable, Authoritative, Charismatic, Courageous, Creative, Exuberant, Flexible thinking, Honest, Leader, Loyal, Modest, Outgoing, Team player, etc.
| |
| | |
| When required by management, standardized surveys help recruiters save time by measuring traits expected for jobs and in candidates. Does one want to know whether a candidate can demonstrate leadership or creativity? The assessment technique will provide the answer. However, what’s really tested is the candidate's understanding of the concept being measured and its impact on their Effective behavior<ref>This is the terminology of GRI’s adaptive profiles, which designates the behaviors effectively expressed by a person and most likely seen by observers. This profile is distinct from the natural and role behaviors measured with the Natural and Role profiles.</ref>. However, the trait model cannot speak to the effort required to sustain it.
| |
|
| |
| Clinicians are trained to understand and diagnose traits observed in clinical practice. Recruiters are also trained to interpret traits, but in work-related behaviors. Some techniques may measure both clinical behaviors and behaviors in the “normal” range, while others focus only on work behaviors and avoid clinical traits, enabling their use by non-clinicians.
| |
| | |
| Trait assessments are typically used before and after development to test a person’s learning of a new concept. Will the change being learned be consistent over time and be reflected in how people act, decide, and emotionally respond? The difference in scores between the first and second measures will tell how much the person learned the concept. However, as with candidates, the technique cannot speak about the effort required to sustain the trait being developed.
| |
| | |
| ==Adaptation Efforts==
| |
| The trait model can inform about how close one is to a targeted destination. In our example with the San Francisco region, the destination could be Union Square. The departure may be anywhere within an 80-mile radius, including the Pacific Ocean, Stockton, Sonoma, or Santa Cruz. The effort required to reach the destination may vary widely, but the model cannot inform about it.
| |
| | |
| In social behaviors, as evidenced by a factor model with measures of adaptation and engagement, successful development will be reflected in how individuals perceive their adaptation<ref>We refer to how people perceive to adapt as the role behavior, represented in the adaptive profile with the Role profile.</ref> and how they ultimately show up<ref>We refer to how people effectively show up as the effective behavior, which combines how people naturally express themselves at flow (the Natural profile), with how they perceive to adapt (the Role profile). It is represented in the adaptive profile with the Effective profile.</ref>.
| |
| | |
| The development will probably not affect the person’s natural behavior<ref>We refer to how people naturally and consistently show up at flow over time as the natural behavior. It is represented in the adaptive profile with the Natural profile.</ref>, which is relatively stable, and if the adaptation is impactful and engaging over time.
| |
| | |
| The trait models, by design, cannot—on the basis of the measure—evidence the efforts candidates or employees make to adapt their behavior to a job, from an emotional, cognitive, and social-behavioral standpoint, nor the amount and quality of support needed for the change to occur.
| |
| | |
| ==Factors versus Traits==
| |
| Factor models are often understood as trait models because factor analysis solved the cumbersome problem of early trait theory by turning thousands of descriptors into a scientific system of measurement. It provided the mathematical evidence that personality includes stable, universal, enduring characteristics rather than temporary states or rigid categories.
| |
|
| |
| Factors are factors; they are not traits. Factors have different attributes and potential than traits. As introduced above, factors are defined by two opposing poles or vectors. Some techniques measure factors but represent them as traits, treating the entire axis as a single trait. But the axis isn't just the high end; it should bring sensitivity to the underlying dimension, something the factor model does. Trait models create a linguistic "shorthand" that makes it seem as if only one end matters.
| |
|
| |
| For example, if the factor is "Patient” vs. "Impatient", being “low" in patience isn't an absence of something; it's the presence of a different, opposing behavior (impatience). As presented above, a factor can be viewed as longitude (East vs. West), while a trait is unipolar and can be viewed as a "fuel tank." You either have a lot of "patience" or very little. In the trait view, "Impatience" isn't a destination; it's just an empty tank of Patience.
| |
|
| |
| A factor model shows that "Low in patience" is its own active state. A trait model simply labels someone as "Not patient," failing to capture the active nature of their behavior. It ignores the "low end" as a distinct quality, thereby losing qualitative data.
| |
| | |
| Modern researchers address the problem of the “low end” by using bipolar factor names (e.g., "Neuroticism vs. Emotional Stability") to ensure that both ends of the continuum have descriptive power. But they often fail to do so because the measures are represented in ways that cannot capture all the meaning along the continuum, which, by design, are antagonistic when they are on both sides, making them difficult to use. In the geolocation analogy, moving West isn't just "failing to move East." It is a specific direction with its own coordinates. Trait models "collapse" the scale, treating the zero point as a void rather than a distinct direction.
| |
|
| |
|
| =Type Model= | | =Type Model= |
| The type model is probably the most comprehensible and widely used, even without any sophisticated formal or statistics-based assessment technique.
| | [[File:Type_San Francisco.png|right|200px]] |
|
| | Type models provide a rough measure of behaviors. With these systems, you are either in an area, as the red area indicated in the illustrationon the right, or you are out. A type provides a very rough understanding of where the journey starts and ends, like going from the East Coast to California, not how it can happen. |
| Types help answer questions about whether a phenomenon exists or doesn’t, such as with hot temperature: Is it hot? Or with competences: is this person competent? Unlike factors and traits, which have continua, types have none. By doing so, the measurement of types disregards anything that doesn’t characterize and is distant from the phenomenon of interest.
| |
|
| |
|
| Like traits, types are easy to understand, probably easier, because how questions are asked and answered come even more naturally and spontaneously with types. Types are easy to learn and use, though they cannot convey nuances as the factor and trait models can.
| | Although the first typologies date back to ancient times, their statistical evaluation is more recent, starting in the 1940s, and their use has grown significantly with the rise of coaching in the late 1980s. Some systems measure only one type (you either have it or do not). Others 4 or up to 20 or 30 types or themes, for what we refer to as multi-typologies. |
|
| |
|
| ==Representation==
| | As for the trait model, the type model doesn’t provide information to understand a person’s development. Unlike trait systems, a type system cannot measure the starting and end points of a development, nor how the development will occur. These are major differences with the trait and factor models. |
| [[File:Type_San Francisco.png|right|250px]]
| |
| Types only provide a rough estimate of a characteristic, whether it be competence or behavior. This is represented by the red square on the map, which indicates that a person is somewhere in the San Francisco region, without knowing the exact location.
| |
| | | |
| If you like to go from A to B on that red map you can’t possibly know, because you don’t have the information. Or if you come from another location, you may end up in the region, including the Pacific Ocean.
| | Uses of type systems typically include coaching, team building, and counselling. Types are not used in recruitment or other uses of the traits and factor models, but only rarely for very rough approximations. |
|
| |
|
| By design, types cannot capture nuances that may be important. This becomes evident when comparing the results of types with those of trait and factor models: those two reveal characteristics that types can’t. If you only use types, you can’t really know until you repeatedly try other techniques, and as in our example, you will land on different locations of the square every time.
| | =Measurement= |
|
| | The preferred technique for measuring social behavior is the questionnaire. However, in our analysis, we had to differentiate the techniques that are forced scenarios from those that are free scenarios. |
| The measures of types suggest that these rough characteristics are likely to persist over time and are attached to people. But that’s far from reality and what everyone can experience on their own. As for competencies, behaviors develop and grow. People adapt, and eventually change, or not. Some behaviors may be engaging; others may be less engaging, in different ways for everyone.
| |
|
| |
|
| Having a picture of whether someone is competent and exhibits a particular behavior type today cannot capture how they will be competent and behave tomorrow and beyond. The type being measured is only a far-from-perfect indicator of competence and behavior at the time of assessment.
| | In the first case, forced-scenarios, the items are propositions that invite the participant to score on a five- to ten-point Likert scale, depending on the technique. In the latter case, the free-scenarios, the technique leaves the participant free to choose the situation or scenario associated with the items to be selected or rejected. The items are typically adjectives. We used to refer to free versus forced-choice techniques, but the distinction using “choice” was limited to a few techniques only (namely, DISC-like techniques). The other expression used was “projective,” which, in psychoanalysis, refers to the unconscious self and to items that people unconsciously describe, such as inkblots. But this distinction was inappropriate in our context. |
| The first impression of being of a type subsists and further misguides consistent development. Types may inform about a person’s modus operandi, but inconsistently, due to their limitations. | |
| | |
| Some systems combine the type and trait models to inform about a person's type’s representativeness and intensity, as traits do, as shown in the example below on the right.
| |
| | | |
| The results are presented on the left as types, and on the right as types + traits. But we see type first and trait second because types, or red dots, are easier to capture and use. It’s no effort for the brain, but also for the immediate deductions made from the types being measured. When combined, the same limitations apply to understanding traits: only one side of the continuum is revealed, and the other is ignored. | | The techniques that have adopted the free scenario strategy, which there are a dozen on the market in the 2020s, and which is the one adopted at GRI, have enabled measuring adaptation and engagement levels in addition to behaviors. It has also helped measure and distinguish between Natural, Role, and Effective behaviors, which other techniques that use the force scenario approach don’t. |
|
| |
|
| [[File:Type_circular representation.png|center|600px]]
| | Interestingly, it is the free-scenario strategy that has helped the factor approach to personality emerge. Then the fixed-scenario took over for measuring the factors being evidenced. The fixed-scenario naturally led to measuring traits and types, and does not take the environment into account the same way that the techniques using the free-scenario strategy do. |
|
| |
|
| ==Measure== | | =Representation= |
| [[File:Type_temperature_simple.png|right|120px]] | | The distinction of three models: Factor, Trait, and Type, has primarily been established based on the representation and use of assessment techniques<ref>This refers to the variable T-REPR of the General Framework. See [[ Assessment_Techniques_-_Characteristics#Assessment_Results | here in this article]] for more details.</ref>. The intrinsic qualities predispose the techniques for what can be done with them. Statistics used are of critical importance. How the techniques are published and distributed, and who uses them for what, is also important to consider. Ultimately, the GRI’s general framework helped uncover that the techniques’ representational quality enables what can be done with it, which falls into the three distinct categories highlighted. |
| A type is a measure that is represented by a square that includes all the elements used to conceptualise the phenomenon of interest by a person or a technique. The phenomenon of focus is the center white square. Others that are not are outside in the outer black square.
| |
| | | |
| In the example on the right, the focus is on answering the question “Is it hot?” The characteristics that qualify as hot are in the square. Others are out. The considerations differ from those of the factor and trait models, which address the development or proximity of the phenomenon.
| | The techniques are used to both measure and represent behaviors, but the two aspects, measurement and representation, need to be considered separately, because a technique may use one model for measurement, but its representation may be based on one or a combination of the other two models. |
| The measure of a type may contain nuances, but this is not of concern for the representation and use of types. In comparison, the factor and trait models will reveal nuances that differ from those of a type assessment, especially from the factor model.
| |
|
| |
|
| As for traits, types can be any individual characteristics one may think of. But unlike traits, types have no intensity. As with traits, factors will reveal how types are expressed and learned, with additional nuances that the type measurement cannot reveal.
| | For instance, some techniques measure factors and are thus factor-based. Whatever scales they use, which may vary greatly between the decile and standard deviation scales, they convert the measures of factors into traits or types and use the results as such, rather than using the factors themselves. The same happens with techniques that measure traits, which are represented as types. |
| | |
| ==Types’ Measurement and Use==
| |
| Types, like traits, are generally measured with forced-choice scenario surveys.
| |
|
| |
|
| Types have been used in personal development and coaching to form a first impression of a person: gauge who they are, and establish a connection. Factor and trait models are used this way, too, each revealing additional precision and capabilities.
| | The techniques are used as they are represented, not for what they measure. Because of the combinations we see among the models, we differentiate them based on how they are both measured and represented, but primarily on how they are represented. |
| When taken at the organizational level, types are used to regroup people into categories, typically a 2x2 matrix, and provide a first understanding of a group’s dynamics and of people who may fit into them.
| |
|
| |
|
| But they lack the nuances needed to, for instance, better characterize what is expected of a person in a job or what happens at a team level. The limitations and imprecision of type measurements prevent making any meaningful, valid deductions at the individual and job levels.
| | =Adaptation= |
| | Adaptation is an important aspect of what the three models deal with, because the object of their measure and representation, behavior, is in ongoing flux. There is constantly something happening out there: a critical or minor event, radical change, disruption, crisis, drama, etc., and also from inside: joy, sadness, health issues, lack of sleep, just a bad or exceptional day, psychological challenges, etc., that constantly ask us to act, the same way actors do on stage. |
| | | |
| Types are used for team building, which, through their far-from-perfect descriptions, enables people to think about behavioral differences. At a minimum, a team-building exercise helps challenge assumptions that focus exclusively on cognitive abilities or other characteristics that are more evident but less relevant to understanding social interactions.
| | Additionally, there are always expectations, including from management or HR, and, at a broader level, from our role in the organization, to behave in a certain way and adapt our behavior accordingly. |
| | | |
| Types tap into people’s curiosity, are easy to start a discussion with, and benefit from what we’ve called art GRI medium effects: the capacity of a technique to help engage people in discussion. Some people may not otherwise share without the type used as a medium, which can be a first great accomplishment.
| | This important aspect of adaptation has repercussions for how one envisions a person’s growth, provides understanding and support, and ultimately, how the person takes on an assignment or a job and performs, including on an emotional level. Development happens differently for everyone, an aspect revealed or not and with varying degrees of accuracy by the three models. |
|
| |
|
| ==Adaptation Efforts== | | =Distinct Benefits of the Factor Model= |
| With social behaviors, types provide an approximate estimate of individual characteristics but, without more nuance, fail to inform what it takes to adapt the behaviors to a job demand, from an emotional, cognitive, and social-behavioral standpoint, nor the amount and quality of support needed for the change to occur.
| | The factor model, with roots dating back to the era of Taylor, has emerged progressively since the 1950s. In social behavior, a major obstacle to their use has been the availability of trait and type models that are easy to understand and use. These two other approaches, however, have drawbacks and limitations that need to be better understood. |
| | | |
| In the map analogy, the location of someone’s behavior is broad. When going somewhere like San Francisco, the departure may be anywhere within the red square on the map. As for the trait model, the departure may be in the Pacific Ocean, in Sonoma or Santa Clara Counties, all of which are 40 to 80 miles distant. Or it may be in San Francisco and even in Union Square, the targeted destination in our example: there is no possible indication of the distance to the target, as with the trait and factor models. You may well already be at the destination, but the model cannot tell and only provides a rough sense that you are around.
| | The factor model offers a nuanced snapshot of individual behaviors, including most preferred and predictable behaviors, the efforts required to adapt, and engage in the job. It also provides a refined understanding of the behaviors expected at the job, team, and organizational levels, which can be represented the same way as with people, helping to see and manage the gaps between the levels. |
| | | |
| ==Factors versus Types==
| | With critical nuances and representations that other models cannot handle, the factor model can better serve executives and managers in developing their social skills, coaches and recruiters to work more effectively with them, and researchers investigating and testing new models that evidence people flow and adaptation efforts. In a similar way, type models have helped coaches and trait models have helped clinicians and recruiters for decades to serve management and organizational needs, but with limited capabilities. |
| Unlike type models, which roughly describe someone’s characteristics in a square, factor models provide their coordinates on the human spectrum.
| |
| When applying factor models to types, people’s behavior doesn't cluster into "types." Instead, they fall on a continuum. Applying a standard deviation scale, most, around two-thirds, fall in the middle of the continuum. The "type" is an arbitrary cut-off on the factor scale, with a divide between the two halves that looks like an empty space rather than being filled with most people.
| |
| | | |
| Type models fail factor analysis because they assume a bimodal distribution of human behavior: people are either in one of two categories, type A or type B. With the geolocation analogy, a type model divides the world into two halves: "The East" and "The West." Anyone at 1° West is grouped with someone at 179° West. A Factor model gives the exact longitude. It recognizes that 1° West is actually much more similar to 1° East than it is to 179° West.
| | The real competing models to the three models, though, are executives’ and managers’ private techniques. Social behavior is the component of personality that’s the most universal, prevalent, intuitive, and subjective, and the one that takes the longest to educate. By bringing unparalleled precision and utility, including at a strategic level, the factor model provides the missing piece for social skills metrics to enhance decision-making and communication. |
| | |
| Some techniques measure dimensions along continuums, but then reduce their representation to types. They do that because it’s easier to think in terms of types than continuums, but it is not appropriate when requiring nuances, as for instance in personal and organizational development.
| |
| | |
| =Summary=
| |
| | |
| The three types of techniques are summarized in the table below.
| |
| | |
| {|class="wikitable" style="margin: auto;"
| |
| ! !! Type !! Trait !! Factor
| |
| |-
| |
| | Representation<br/>Locating Social Behavior || [[File:Type_San Francisco.png|center|200px]]Behaviors are roughly measured. || [[File:Trait_San Francisco.png|center|200px]]The distance to a targeted behavior is measured. || [[File:Factor_San Francisco.png|center|200px]]Behaviors are precisely measured, including the distance to an objective.
| |
| |-
| |
| | Measurement || [[File:Type_Representation.png|center|100px]]Measure a rough location. Being in or out. || [[File:Trait_Representation.png|center|100px]]Measure the distance to the target, from "around." || [[File:Factor_Representation.png|center|200px]]Reveals precise location with nuances.
| |
| |-
| |
| | Adapting Behavior, Development || A rough estimate of where the adaptation starts and ends, but not how it happens, nor the effort it takes to sustain the effort. || Understanding how distant the development is at the start and the end, but not how it happens, nor the effort it takes to sustain the effort. || Understanding where the development starts, where it ends, how it happens, and the effort it takes to sustain it.
| |
| |-
| |
| | Typical Usage || Coaching, team building, counselling || Recruitment, vocational guidance, and other uses of types || Management, leadership, organizational development, and other uses of types and traits
| |
| |-
| |
| | Typically Measured || Up to a few dozen types || Up to a few dozen traits || Up to seven factors
| |
| |-
| |
| | Users || Coaches || Recruiters, clinicians, and coaches || Executives, managers, recruiters, clinicians and coaches
| |
| |-
| |
| | Emergence/Use || 1940s/1980s || Late 1800s/1920s || 1950s/2010s
| |
| |}
| |
|
| |
|
| The factor approach, with roots in research dating back to the era of Taylor, has progressively emerged since the 1950s. With social behavior, a major obstacle to their use has been the availability of trait and types models that are more immediate to understand and use but however have drawbacks that this article discusses.
| |
|
| |
| The factor model uniquely serves management and organizational development by offering a nuanced snapshot of individual behaviors, including most preferred and predictable behaviors, the efforts required to adapt, and engage in the job.
| |
|
| |
| It also provides a refined understanding of the behaviors expected at the job, team, and organizational levels, which can be represented the same way as with people, helping to see and fill the gaps between the levels.
| |
|
| |
| With critical nuances and representations that other models cannot handle, the factor model can better serve executives and managers in developing their social skills, coaches and recruiters to work more effectively with them, and researchers investigating and testing new models. In a similar way, type models have helped coaches, and trait models clinicians and recruiters for decades to serve management and organizational needs, but with limited capabilities.
| |
|
| |
| The real competing models to the three models, though, are executives’ and managers’ private techniques. Social behavior is the component of personality that’s the most universal, prevalent, intuitive, and subjective, and the one that takes the longest to educate. By bringing unparalleled precision and utility, including at a strategic level, the factor model provides the missing piece for social skills metrics to enhance decision making and communication.
| |
|
| |
|
| =Notes= | | =Notes= |
Introduction
For decades, researchers and the personality assessment industry have studied and refined various For decades, researchers and the personality assessment industry have studied and refined various methods for measuring personality in organizations. These methods are used for hiring, personal development, coaching, and more, including in clinical settings. We at GRI have reviewed these methods by analyzing their effects and benefits on individual and organizational performance, how they were developed, and by studying their users: traditional users like human resources managers, recruiters, coaches, and clinicians, as well as managers and executives.
In social matters, unlike in technology, change happens slowly. As for assessment techniques, particularly with personality, as their history shows, understanding their true potential takes time[1]. This extended process may be due to a combination of factors: the abstract nature of the concepts being measured, the effort needed to learn these techniques, and the emotional attachment people develop toward them.
Assessment techniques can reveal their real capabilities through actual cases that genuinely matter. These cases usually come from organizations that care not only about performance in finance and production but also about how these outcomes are delivered from behavioral, cognitive, and emotional perspectives.
Using both qualitative and quantitative methods has helped us gain a detailed understanding of assessment techniques' capabilities and limitations: what they truly measure, how these measures are represented and used, and the questions they seek to answer. These topics are discussed below.
The Three Models
Three categories of models clearly emerge from analyzing the assessment techniques used in organizations and research labs, taking into account their characteristics and uses. 28 criteria have allowed us to differentiate assessment techniques that are particularly applicable to social behavior[2]. The three categories appear as the biggest differentiators from both conceptual and practical standpoints.
The three models presented below offer different understandings of how people function from behavioral, emotional, and cognitive standpoints. Although each can be useful for addressing questions about people, they also have limitations. Seeing people in context, or not, and with varying nuances, the three models ultimately carry a different vision of how people can be understood, how they possibly show up, grow, and perform.
The analogy of a two-dimensional map is often used to illustrate the importance of accurately understanding or locating behaviors. We also use it to explain the differences among the three models, to understand the potential challenges and efforts required to move from point A to B. Above all, those models speak to action and, at least partially, account for how action unfolds as we navigate the map.
The three models are the following. Each is detailed in a separate article. They are briefly presented below.
| Model |
Description
|
| Factor |
Behaviors are precisely measured by their coordinates on the map. The distance and effort required to cover distances on the map can be anticipated.
|
| Trait |
Behaviors are estimated by recouping small areas on the map. One knows we are more or less distant from it. Distances to be covered can only be roughly approximated.
|
| Type |
Being regrouped in broad categories, behaviors on the map appear as large areas, making it impossible to predict the distances to cover.
|
Factor Model
With the factor model, as the illustration on the right shows, behaviors are precisely measured, like knowing latitude and longitude on a map. When moving from one point to another on the map, it is possible to anticipate the route, calculate the distance, and select the best way.
The factor approach emerged in the 1950s but did not gain acceptance until research from the 1970s to the 1990s promoted a limited number of factors to explain behavioral traits. Systems based on factors work with a limited number of them. We believe at GRI that four is the optimal number. Some models in research work with two, three, five, and up to seven factors.
When used to understand behavior adaptation and individual growth, the factor approach identifies where development starts, where it ends, how it happens, and the effort required to sustain the adaptation.
Potential users for the factor model are executives, managers, and other users of types and traits (see below). Applications of the factor model in organizations include management, leadership, and organizational development, as well as uses related to types and traits (see below).
Trait Model
As illustrated by the red circle on the right, measures from a trait system indicate how far one is from a target, but that can be from different locations that we don’t know. Additionally, the route and time between locations can only be a rough but impractical estimate; we cannot know either.
The trait model was the first to be used in psychiatry and clinical practice in the late 1800s. It was soon employed for large-scale recruitments in the 1920s, and later for executive search. The number of traits that a system from this model can measure is unlimited. It can be as low as one trait in clinical applications. In organizational applications, one trait can be leadership, creativity, or honesty. In recruitment, the number may be as high as 30. This number can easily double or triple with batteries of tests.
Trait models are not built to understand a person’s development but measure the starting and end points of a development, not how it will occur. That’s another major difference with the factor model.
Potential users of trait systems include clinicians, recruiters, and coaches, but not executives and managers directly. Trait systems are used in organizations for recruitment and vocational guidance. They are also used like type systems (see below), but rarely cover the applications seen with the factor model (see above).
Type Model
Type models provide a rough measure of behaviors. With these systems, you are either in an area, as the red area indicated in the illustrationon the right, or you are out. A type provides a very rough understanding of where the journey starts and ends, like going from the East Coast to California, not how it can happen.
Although the first typologies date back to ancient times, their statistical evaluation is more recent, starting in the 1940s, and their use has grown significantly with the rise of coaching in the late 1980s. Some systems measure only one type (you either have it or do not). Others 4 or up to 20 or 30 types or themes, for what we refer to as multi-typologies.
As for the trait model, the type model doesn’t provide information to understand a person’s development. Unlike trait systems, a type system cannot measure the starting and end points of a development, nor how the development will occur. These are major differences with the trait and factor models.
Uses of type systems typically include coaching, team building, and counselling. Types are not used in recruitment or other uses of the traits and factor models, but only rarely for very rough approximations.
Measurement
The preferred technique for measuring social behavior is the questionnaire. However, in our analysis, we had to differentiate the techniques that are forced scenarios from those that are free scenarios.
In the first case, forced-scenarios, the items are propositions that invite the participant to score on a five- to ten-point Likert scale, depending on the technique. In the latter case, the free-scenarios, the technique leaves the participant free to choose the situation or scenario associated with the items to be selected or rejected. The items are typically adjectives. We used to refer to free versus forced-choice techniques, but the distinction using “choice” was limited to a few techniques only (namely, DISC-like techniques). The other expression used was “projective,” which, in psychoanalysis, refers to the unconscious self and to items that people unconsciously describe, such as inkblots. But this distinction was inappropriate in our context.
The techniques that have adopted the free scenario strategy, which there are a dozen on the market in the 2020s, and which is the one adopted at GRI, have enabled measuring adaptation and engagement levels in addition to behaviors. It has also helped measure and distinguish between Natural, Role, and Effective behaviors, which other techniques that use the force scenario approach don’t.
Interestingly, it is the free-scenario strategy that has helped the factor approach to personality emerge. Then the fixed-scenario took over for measuring the factors being evidenced. The fixed-scenario naturally led to measuring traits and types, and does not take the environment into account the same way that the techniques using the free-scenario strategy do.
Representation
The distinction of three models: Factor, Trait, and Type, has primarily been established based on the representation and use of assessment techniques[3]. The intrinsic qualities predispose the techniques for what can be done with them. Statistics used are of critical importance. How the techniques are published and distributed, and who uses them for what, is also important to consider. Ultimately, the GRI’s general framework helped uncover that the techniques’ representational quality enables what can be done with it, which falls into the three distinct categories highlighted.
The techniques are used to both measure and represent behaviors, but the two aspects, measurement and representation, need to be considered separately, because a technique may use one model for measurement, but its representation may be based on one or a combination of the other two models.
For instance, some techniques measure factors and are thus factor-based. Whatever scales they use, which may vary greatly between the decile and standard deviation scales, they convert the measures of factors into traits or types and use the results as such, rather than using the factors themselves. The same happens with techniques that measure traits, which are represented as types.
The techniques are used as they are represented, not for what they measure. Because of the combinations we see among the models, we differentiate them based on how they are both measured and represented, but primarily on how they are represented.
Adaptation
Adaptation is an important aspect of what the three models deal with, because the object of their measure and representation, behavior, is in ongoing flux. There is constantly something happening out there: a critical or minor event, radical change, disruption, crisis, drama, etc., and also from inside: joy, sadness, health issues, lack of sleep, just a bad or exceptional day, psychological challenges, etc., that constantly ask us to act, the same way actors do on stage.
Additionally, there are always expectations, including from management or HR, and, at a broader level, from our role in the organization, to behave in a certain way and adapt our behavior accordingly.
This important aspect of adaptation has repercussions for how one envisions a person’s growth, provides understanding and support, and ultimately, how the person takes on an assignment or a job and performs, including on an emotional level. Development happens differently for everyone, an aspect revealed or not and with varying degrees of accuracy by the three models.
Distinct Benefits of the Factor Model
The factor model, with roots dating back to the era of Taylor, has emerged progressively since the 1950s. In social behavior, a major obstacle to their use has been the availability of trait and type models that are easy to understand and use. These two other approaches, however, have drawbacks and limitations that need to be better understood.
The factor model offers a nuanced snapshot of individual behaviors, including most preferred and predictable behaviors, the efforts required to adapt, and engage in the job. It also provides a refined understanding of the behaviors expected at the job, team, and organizational levels, which can be represented the same way as with people, helping to see and manage the gaps between the levels.
With critical nuances and representations that other models cannot handle, the factor model can better serve executives and managers in developing their social skills, coaches and recruiters to work more effectively with them, and researchers investigating and testing new models that evidence people flow and adaptation efforts. In a similar way, type models have helped coaches and trait models have helped clinicians and recruiters for decades to serve management and organizational needs, but with limited capabilities.
The real competing models to the three models, though, are executives’ and managers’ private techniques. Social behavior is the component of personality that’s the most universal, prevalent, intuitive, and subjective, and the one that takes the longest to educate. By bringing unparalleled precision and utility, including at a strategic level, the factor model provides the missing piece for social skills metrics to enhance decision-making and communication.
Notes
