In the process of interacting with each other, people need mutual understanding. If there are not enough facts to explain the behavior of another person, then observers tend to attribute different motives. The same applies to the object of discussion itself: he is also trying to find the reasons for his result. This phenomenon is called causal attribution - attributing causes without knowing for sure the content of what happened. It began to be studied in Western social psychology. Haider is considered to be the founder.
Casual attribution in psychology. Attribution Examples
This phenomenon exists because everyone wants to see the whole picture, to imagine all the events. But the problem is that the facts are not always known. And then the person begins to finish drawing, to think out the picture, bringing it to a logical conclusion. This process is carried out in accordance with existing life experience. Psychology has noted the diverse reactions of society to stereotypical and deviant behaviors . Let's look at an example.
The students are waiting for a new teacher to teach them history. If you ask them to describe their history teacher, then most likely the classes will be boring and uninteresting. And if you introduce them to another teacher, having previously described his teaching style (he uses visual models, arranges skits, does everything to make the lessons interesting), then the opinion about the person will be non-standard, different from the common habitual judgment.
CoMagic Attribution Models
For most types of business, standard models are suitable - CoMagic has them all. If you need everything to be simple and clear to make the right and quick decisions, in your personal account you will find the following report: “Attribution Models”. It allows you to choose the model that suits your business and change it in real time. This way you can immediately see how the overall picture of channel efficiency is changing.
When choosing the model that suits you best, you can start from the standard Last Click model. If the selected model counts fewer conversions than Last Interaction, then the channel is overvalued under the standard model. If the model we are studying shows more conversions, most likely, this channel was underestimated.
Svyatoslav Vasiliev, Product Director
Oddly enough, a large number of companies still analyze advertising only using the last click model. This is bad not only because in many businesses with this model organic and direct visits come to the fore, but also because the sources that lead to previous touches of visitors who converted are washed out.
For those who understand the importance of tracking all sources along the path to a sale, we have released Attribution Models and Assisted Conversions. Together with our end-to-end analytics, they allow you to see from what sources the first acquaintance of a future buyer with your website occurs. Or remove all direct calls from analytics so that they do not push aside those sources that you can influence. Also in our report you can analyze keys from all advertising systems simultaneously to evaluate the overall contribution of a key regardless of the advertising system. If you're not sure which attribution model best describes the value of ad sources' contributions, you can use the Assisted Conversions report to ensure you don't accidentally disable an ad source that's not driving conversions.
A good question is which lookback window to choose when using attribution models. To do this, you need to know your first-touch conversion cycle. In CoMagic, the lookback window can be set to 180 days, which means that we will look for referrals from sources for a visitor six months before they convert. If you need a lower value, you can change it in the report settings.
It is important to understand the main thing: whatever approach you choose to assess the significance of sources on the path to conversion, in any case it will be better than evaluating advertising only according to the last click model and not wondering about the length of the path to conversion and the significance of advertising sources on other places along the way, except the last one. And to understand what works best for your specific business, there is only one tool - experiments.
Types of causal attribution
Depending on the angle from which the situation is viewed, the resulting result is determined. The following types are distinguished :
- Personal attribution. Attributing the causes of failure directly to the individual;
- Circumstantial. Blaming the circumstances;
- Object. The reason is in the object itself.
It is interesting that a person’s position determines the direction of his thinking. The participant himself most often blames the circumstances. The observer sees the motive for failure in the individual (participant). This happens because neither one nor the other imagines a completely plausible picture. It turns out that attribution is a subjective, and therefore often erroneous, opinion.
One more example. A shy guy finally decided to meet a girl. I thought about everything in advance, even rehearsed my speech. In general, he also raised his self-esteem. He meets her on the street, but for some reason she refuses the opportunity to meet her. The guy immediately builds all sorts of hypotheses. He thinks: “maybe it’s just me, maybe she doesn’t like me; maybe she’s just not in the mood,” etc. These thoughts can be either separately or come one after another.
At the same time, a correct understanding of the reasons for a person’s actions is very important for maintaining relationships between people . Contrived motives for behavior can be very different from real motives. But it turns out that sometimes a person cannot ask, clarify some points and therefore is forced to use his imagination.
Standard attribution models
By default, Google Analytics offers us to use several models: First Click, Last Click, Last Non Direct Click. Over time, additional models appeared:
- “Linear”—each point along the conversion path is assigned equal shares of the conversion value (25% each).
- “Taking into account the recency of interactions” - the largest share of value goes to touchpoints that are closest in time to a sale or conversion.
- "Position-based" - 40% of the value is assigned to the first and last interactions, and the remaining 20% is equally distributed among the others.
However, all of these models have their drawbacks. For example, First Click and Last Click take into account the first and last channel, while all other channels remain underestimated. The Time Decay Model works like this: the closer a channel is to the final conversion, the higher its value. However, this may not be entirely correct, because channels in longer but effective chains will look worse than the same channels in short chains. Moreover, the importance of the first or last touch often greatly depends on the scope of a particular business, and it may not be entirely correct to underestimate them either.
What other attribution models can there be? Konstantin Yurevich told us about an attribution model developed based on machine learning.
Goals and results of causal attribution research
The goal of research into the mechanisms of causal attribution is to increase the effectiveness of interaction between people and the effectiveness of personal growth. The first presupposes the most accurate determination of the motives of certain actions. And the second shows options for influencing motivation, activity, emotions, etc. What most fully helps to understand the study of this phenomenon is the indication of the moment of assigning or accepting responsibility for specific actions. And a comprehensive consideration of the current result. That is, the goal of research is to find an accurate definition of the actual motives of behavior .
Errors of perception
It is known that a person treats himself more softly when assessing than other strangers. A person attributes someone’s successes and his own failures to situational attribution. But, describing other people's failures and his own successes, he turns to personal attribution. In these cases, a person tends to consider the cause of what happened to be either the prevailing circumstances or the person himself, according to the final result.
Usually a person explains success by his hard work, willpower, and his uniqueness. But failure is always associated with the situation. And if you analyze the actions of another person, then all of the above applies in reverse order. If a person achieves success, it is because the circumstances have developed that way. And if he failed, then he himself is to blame . And few people think differently. Few will pay attention to the situation and focus on it. After all, if you explain the result of a person’s activity in a different way, then this means recognizing it at your level, or even better. This means comparing him to yourself.
Therefore, people tend to protect their self-esteem in this way. It is easier to blame the circumstances, the object of the action, than to force yourself to work, to improve yourself. Causal attribution is applicable everywhere: in everyday life, at work, in relationships. And this principle of opposites operates everywhere.
Attribution Theory
Explains how and why many make ordinary inferences that mediate between the perception of events and the response to them.
Inferences are judgments that influence behavior, feeling, and the conclusions we make about events. Conventional inferences involve errors in judgment.
According to attribution theory, the behavior of other people is determined by the qualities of the person and the situation in which they find themselves.
However, there is a general tendency to link the causes of behavior to the internal characters of a person.
This is the fundamental attribution error.
Conditions for the occurrence of fun. Attribution error.
1) If we believe that others have a choice in a particular situation, then in most cases, we will regard their actions as intentional.
2) If a person’s actions concern us personally, then we will most likely draw a conclusion that the person has internal motives to act one way and not another.
3) People also make inferences about their own behavior. They often attribute success to themselves, and blame external factors for their success.
When the reasons for another person’s behavior are unknown, a means of explaining behavior (and a social phenomenon in general) is attribution, i.e., a kind of completion of information is carried out. Attribution is the process of perception by a “naive psychologist” of the causes of behavior and its results, allowing a person to give meaning to the environment. introduced the term and studied causal attribution . According to causal attribution
A person's perception of the behavior of others is largely determined by what he believes to be the cause of that behavior. The causes of behavior are usually explained by individual (personal) characteristics or the situation in which the behavior was manifested, or a combination of both.
Dispositional (personal, internal, internal) attribution emphasizes certain aspects (abilities, skills, motives) of the individual, and situational (external, external) attribution emphasizes the influence of the external environment on behavior (being late for work can be explained by snow drifts on the road). Thus, we evaluate the behavior of others based on our perceived motives and intentions. Heider's work provided the conceptual basis for a more general approach known as attribution theory.
Modern theories of attribution, dealing with problems of social perception, actually try to explain (understand, know) how they attribute characteristics and qualities to other people.
The act of attribution is the attribution or endowment of some characteristics (or traits, emotions, motives, etc.) to oneself or another person. The term represents not so much a formal theory as a general approach in social and personality psychology in which behavior is studied in the light of this concept. Using the Gestalt psychology
proposition that information acquired through the observer's past experiences plays an important role in the processing of new data, attribution theories argue that in social situations the following sequence is observed: a person observes the behavior of another person, making a logical inference about the intentions of this a person, based on the perceived data, and then attributes to him some hidden motives that are consistent with this behavior.
There are many variations on this theme, including self-perception theory,
which views a person's self-image within this theoretical approach.
The fundamental attribution error is the tendency of people to ignore situational reasons for actions and their results in favor of dispositional (personal) ones. This error is not absolute (it is not universal, it does not always appear, not under all circumstances, it can be taught to recognize and eliminate) (Fig. 2.4.1). Conditions for fundamental attribution error to occur
are [15]:
• "False consent"
- this is an overestimation of the typicality of one’s behavior (one’s feelings, beliefs, convictions), expressed in the fact that the observer considers his point of view to be the only correct (“normal”), which should be characteristic of all people, and any deviation from it is associated with the personality of the observed ( figure).
• “Unequal opportunities”
– this is the failure to take into account the role position of the acting (observed) person. Each person plays many roles, and some of their roles make it easier to express themselves and bring out positive qualities. It is this mechanism that is activated during attribution in manager–subordinate situations.
• “Ignoring the informational value of what did not happen.”
Information about what “didn’t happen”—about what a person “didn’t do”—may be the basis for assessing behavior, but it is this information that is often omitted, since the observer superficially perceives only “what happened.”
• “More confidence in facts than in judgments.”
The first glance is always directed to a “more salient” fact - to the person (“figure”), and the situation (“background”) still needs to be determined. This is where the “figure-ground” perception focusing mechanism comes into play.
• “Ease of constructing spurious correlations.”
This phenomenon forms the basis of
implicit theories of personality
and consists in the fact that a naive observer arbitrarily connects any two personality traits as necessarily accompanying each other. Particularly often, an arbitrary combination of external traits and psychological properties is carried out, which speeds up and simplifies the attribution process (for example, all overweight people are kind, all short male managers are power-hungry, etc.).
Rice. 2.4.1. Fundamental attribution error
•
Some
social norms in different cultures
create a tendency towards a certain type of attribution (Western individualism - towards personal attribution, and Eastern collectivism - towards situational attribution).
• “Locus of control”[17]
(internal or external) influences how people (internals or externals) “see the world,” in particular the type of attribution they prefer: internals more often use personal attribution, and externals use situational attribution.
• Bias of the “doer-observer”
manifests itself in a significant difference in their perceptual positions, which are expressed in the fact that they: a) have different levels of information: the observer knows little about the situation, and the actor acts on the basis of his experience in a situation developing over time; b) have a different “angle of view” on what is observed, that is, they have a different perceptual focus: for the observer the background is the situation, and the actor is the figure, and for the actor the situation is the figure.
Attributions and errors in this process occur in predictable cases. Kelly's research on the attribution process identified three factors that influence the attribution process and determine behavior:
• consensus (unanimity) –
how many people behave in the same way as a certain individual; we tend to attribute personal motives to unusual behavior and situational motives to behavior common to many;
• intentionality –
how thoughtful or unusual the individual’s behavior is in a given situation; we tend to attribute personal motives to deliberate, calculated behavior, and situational reasons and motives to unusual, non-standard behavior;
• subsequence -
How consistent is an individual’s behavior throughout time and in terms of the sum of all actions. We tend to attribute consistent behavior of an individual to personal motives, and situational reasons that determined the behavior to isolated cases of behavior.
Attribution of the cause or motives of a given event to personal or situational factors also depends on the individual’s point of view, the appropriateness and possibility of choosing a given behavior, the nature of the consequences, the individual’s propensity for responsibility, and the individual’s personal concept (self-concept).
Any attributional process begins with a person’s motivation to understand (know) the causes and consequences of the actions of other people (and himself), and then predict their behavior in the future, i.e., understand the meaning of human relationships. Motivational attribution errors are represented by various kinds of “defenses” (protection of one’s own Ego, “I”-concept), biases (tendency to see oneself in a more favorable light, inflated self-esteem) and are caused by a person’s subjective interpretation of social reality, which inevitably includes bias (bias, partiality) many judgments.
Significant credit for the development of this problem belongs to B. Weiner. Its focus is on the locus (focus) of causation—the ways in which causes are attributed in situations of success and failure. He proposed to consider three dimensions in each cause: internal-external; stable-unstable; controlled-uncontrolled. Various combinations of these dimensions give eight models (possible combinations of causes) and are a motivated assessment (success or failure) of an event. When analyzing the performance of a task, four classic factors should be taken into account: ability, effort, difficulty of the task, success, on which, according to Weiner, the nature of any action depends (Table 2.4.1).
Table 2.4.1
