What psychological theories are pseudosciences

Science and pseudoscience in psychology. Author: Siegfried Macho

Transcript

1 Science and pseudoscience in psychology Author: Siegfried Macho

2 Table of Contents i Table of Contents 1.

3

4 Table of Contents iii 4.

5 Chapter 1: Introduction 1 1. Introduction: The problem of differentiating between science and pseudoscience If one looks at the history of various scientific disciplines, one encounters numerous pseudoscientific theories and practices that lead a kind of independent existence, parallel to the actual disciplines. Typical examples are: alchemy (chemistry), astrology (astronomy), creationism (Darwinism / biology), quackery (medicine). Interestingly, there are always cases of prominent scientists who, in addition to their scientific work, deal extensively with pseudoscientific theories and practices. Well-known examples of this are Isaac Newton (), who probably devoted more time to his alchemical studies than to his scientific work (Westfall, 1980), or Johannes Kepler (), who is said to have created horoscopes for Wallenstein. In addition to medicine, psychology as a scientific discipline provides a rich breeding ground for pseudoscientific theories and practices. Pseudoscientific activities in the context of psychology range from activities of certain sects (e.g. Dianetics of Scientologists) to New Age therapies (e.g. rebirthing) to dubious diagnostic methods (e.g. graphology). Comment Psychology as a scientific discipline is viewed as a branch of the natural sciences, since its theories serve to explain and predict phenomena in the real world and can and must therefore be subjected to empirical testing. In contrast, humanities disciplines such as philosophy and pure mathematics do not make any empirically testable statements and can therefore be operated in an "armchair" (which in no way means that they are simpler). Now the following question immediately arises: Question: How can science be distinguished from pseudoscience? The answer to this question is not only of theoretical but also of eminently practical importance. So try in the US

6 Chapter 1: Introduction 2 again and again representatives of the so-called »intelligent design« a further development of creationism (see Kitcher, 2009) to legally enforce recognition or equality of this teaching with the Darwinian theory of evolution. In order to ward off such attempts, it must be shown that the doctrine of "intelligent design" is a pseudoscience. There are a number of indicators for the existence of pseudoscientific theories and / or practices. Method 1-1 summarizes the most important. Method 1-1: Indicators of pseudoscientific practices (Lawson, 2007): 1. Inaccurate, scientific-sounding language: Use of scientific-sounding expressions such as "holistic", "energy", "rhythms", "radiate", "fields", etc. 2. Reference to dubious, idiosyncratic practices: Procedures are only effective if they are carried out in a very specific way, e.g. Medicines that are only effective when shaken in a certain way. Usually only the "initiated" are able to carry out these procedures correctly. 3. Subjective personal experience as the main evidence: Effects of procedures are not determined by controlled studies, but are based on testimonies from individuals. These are often attached to advertisements for a drug or a method as letters of commitment or recommendation. 4. Pretending to have solid empirical evidence: Often, reference is made to a long list of scientific publications, most of which have appeared in some questionable journals. This creates the impression that there is converging or cumulative evidence. Note: There are, however, cases where extremely renowned journals published such questionable results, e.g. Nature, Psychological Bulletin and Journal of Personality and Social Psychology.

7 Chapter 1: Introduction 3 5. Reference to dubious authorities: Often some authorities are cited who have never been scientifically active in the field or who have no in-depth knowledge of the field. Example: Einstein says: We only use 10% of our brain. «Note: Nobody knows whether Einstein ever made any claims in this direction (The research assistants at the Albert Einstein Archive could not find any clues [Beyerstein, 2007, page 54]) ). 6. Reference to old wisdom that has apparently been lost: It is often pointed out that a method is an ancient wisdom of the people, but which has been lost over the centuries or has been handed down by some indigenous peoples. For example: the wisdom of the ancient Chinese or the knowledge of the Indian chief "Sitting Bull". 7. Statements about relationships that are in sharp contradiction to our basic physical knowledge: Statements are made that are in sharp contrast to the physical knowledge known to us. Example: water takes on the properties of every substance that has ever been in it. Therefore, the substance continues to work, even if it can no longer be detected by physico-chemical means. 8. Reversal of the burden of proof: Instead of providing evidence that an entity or an effect exists, the opposite is required of the counterpart to provide evidence that said entity or effect does not exist. E.g .: representatives of religions point out that one cannot prove the non-existence of God. 9. Skirmishes of retreat instead of progress: Good scientific theories are characterized by the fact that they lead to new empirical testable predictions and to new problems. Pseudoscientific theories do not make testable predictions. Instead, you are constantly fighting retreat, in the form of reinterpretations or the partial abandonment of positions (without giving up the core messages).

8 Chapter 1: Introduction 4 Example: The development of creationism towards "intelligent design" is a typical example: Here old positions have been abandoned (namely that God created the earth about 6000 years ago. The central position that man ( at least partially) created by God is retained, but this is obscured by speaking of an "intelligence" that partially interferes with evolution (Kitcher, 2009) The points listed in Method 1-1 speak for themselves and There are numerous examples of pseudoscientific practices in medicine and psychology to which one or more of the indicators listed apply (Exercise 1-1). On the other hand, there have been numerous attempts to establish criteria for scientificity. Method 1-2 presents one Method 1-2: A selection of possible criteria for identifying science: A number of possible characteristics have been suggested n, by which science supposedly distinguishes itself. Here are a few: 1. Science uses clearly defined and undisputed concepts. 2. The theoretical entities postulated or discovered by the sciences, such as forces, fields, representations, cognitive processes, etc., are undoubtedly real existing things on which the observable phenomena are based. 3. Most models of exact science correctly depict reality. 4. Science is getting closer and closer to the truth. 5. Theories that have been falsified by empirical data are excluded from the canon of scientific theories. 6. Science is based on the application of sound methods that have been proven to be reliable. Some of the items listed in Method 1-2 are often found in conversations with laypeople about the nature of science. Let us come back to the question asked above about the demarcation between science and pseudoscience. The two methods seem to bring us closer to solving this problem, as they seem to provide us with clear criteria for identifying scientific and pseudoscientific practices. Unfortunately, this approach is not effective for the following reasons:

9 Chapter 1: Introduction 5 None of the criteria listed in method 1-2 are used to identify scientific activity. In fact, there is still no undisputed criterion that makes it possible to separate scientific practice from other types of activities. There are border areas or gray areas for which it is unclear whether it is a science or rather a pseudo or protoscience (i.e. a discipline that is not yet recognized as a full science). Psychoanalysis is a good example of this. Scientific research that ventures into new areas often has pseudoscientific features. The best example of this is offered by modern theoretical physics, in which some constructs such as extra dimensions, strings, or supersymmetries are postulated, of which it is unclear whether they have anything to do with reality (see e.g. Randall, 2012; Susskind, 2010). The fact that there are no stringent criteria for separating science from other activities has led a number of philosophers of science to postulate a kind of equation between science and myth (Quine 1951; Feyerabend, 1995; Hübner, 1985). Comment In fairness it should be added that there are clear differences between the three authors in terms of their critical attitude towards science. Quine is certainly less critical of science (or even hostile) than the other two authors. This is proven by the following comparison of quotes from the work of Quine and Feyerabend: Physical objects are conceptually imported into the situation as convenient intermediaries not by definition in terms of experience, but simply as irreducible posits comparable, epistemologically, to the gods of Homer. Let me interject that for my part I do, qua lay physicist, believe in physical objects and not in Homer's gods; and I consider it a scientific error to believe otherwise. But in point of epistemological footing the physical objects and the gods differ only in degree and not in kind. Both sorts of entities enter our conception only as cultural posits. The myth of physical objects is epistemologically superior to most in that it has proved more efficacious than other myths as a device for working a manageable structure into the flux of experience (Quine, 1951). In Feyerabend (1995, page 385), however, one reads:

10 Chapter 1: Introduction 6 So there is no clear difference between myth and scientific theories. Science is one of the many forms of life that humankind has evolved, and it is not necessarily the best. It is loud, cheeky, expensive and attracts attention. In principle, however, it is only superior in the eyes of those who have already taken a certain position and who accept science without ever having examined its advantages and weaknesses. Kurt Hübner occupies a middle position here, although over time he tended more and more to the position of Paul Feyerabend. Something cannot be right about a radical equation between myth and science as seen by Feyerabend (1995) and also by Hübner (1985), because it is obvious that the scientific predictions in terms of accuracy and detail are far from those of all mythological systems and also those of everyday understanding surpasses. Furthermore, the knowledge gained from scientific research leads to technological advances that have fundamentally changed the life of mankind [compare the last sentence in the above quotation from Quine (1951)]. Equating science with myth, pseudoscience, etc. also involves considerable dangers. It can namely lead to certain scientific standards that are relevant to the progress of a discipline being weakened. The classic example of this is (evidence-based) medicine. In addition to remedies whose effects have been scientifically tested, there are not only numerous techniques and remedies from the field of alternative (or complementary) medicine, the effectiveness of which is extremely controversial, but there are also numerous attempts to meet the criteria for testing the effectiveness of To soften therapies or drugs. However, it is precisely on these strict methods that the success of evidence-based medicine is based (see Ernst & Singh, 2009). We will go into more detail later on this problem, which also affects clinical-psychological practice. Let us summarize what has been said so far: 1. There is no clear criterion or procedure that would allow scientific theories and practices to be clearly distinguished from other activities. Science is by no means immune to "pseudoscientific challenges," which explains why dubious work is occasionally published in reputable journals. 2. On the other hand, science has turned out to be the most successful undertaking in human history to explain and predict (and thus control and manipulate) events. It is fair to say that everyone else

11 Chapter 1: Introduction 7 whose attempts such as everyday explanation, magic, myth, religion, etc. have shown themselves to be clearly inferior to science. These two statements seem slightly contradicting at first glance, then if there are no clear criteria for scientific activity to be carried out by others, why is the latter so successful? This is because, like most categories, science can be viewed as a fuzzy category. However, it is characterized by the increased occurrence of certain "positive" properties. There are two above all to be mentioned here: 1. The tendency or requirement to put all assumptions and evidence on the table so that they can be subjected to the general criticism of the research community. 2. A continuous criticism from which nothing is excluded. These two aspects taken together have been considered by many to be the defining characteristic of science. This is not entirely correct insofar as people think and act (or at least try to) critically in everyday life. On the other hand, the "mills of criticism" often grind rather slowly, even in science. As we shall see, this also applies to psychological research. Comment on the fuzzy categories; As indicated above, science can be described as a fuzzy category that cannot be distinguished from other human activities by clear criteria. The idea of ​​fuzzy categories was emphasized by Wittgenstein (1953). He looks at different types of games (e.g. chess, or a child throws a ball against a wall) and asks what the common characteristic of all these activities is. He concludes that there is no such common denominator. Rather, categories are characterized by so-called family similarities. This means that the members of a category are similar to one another, but this does not rule out that two different specimens at different poles of the category can be very dissimilar. This idea also became the dominant view within psychology (see e.g. Rosch & Mervis, 1975). The following presentation therefore deals with the following questions: Questions: 1. How does psychology work as a science?

12 Chapter 1: Introduction 8 2. How do common pseudo and protoscientific practices work and how do they differ from scientific practices? 3. What are the reasons for the widespread use of pseudoscientific practices in the various areas? In order to answer these questions, I will proceed as follows: I will begin by presenting various conceptions of science and show how psychology as a scientific discipline fits into these proposed schemes. It addresses the problem of scientific explanation and examines what constitutes a good psychological explanation. Theoretical dynamics, i.e. the formation and rejection of psychological theories and research programs, is examined. We look at various pseudoscientific practices and examine how well they meet the criteria listed in Method 1-1. We turn to the question of central psychological mechanisms of belief in the success of pseudoscientific practices. 1.1 Exercises for Chapter 1 Exercise 1-1: Discuss a pseudoscientific method / theory you are familiar with using the given indicators. Which of the indicators are applicable to this method and which are not? Exercise 1-2: Discuss the criteria given in Method 1-2. Give examples of whether or not the various criteria were met.

13 Chapter 2: Concepts of psychology as a science 9 2. Concepts of psychology as a science The different concepts of science can be divided into 3 broad categories, with the view of Karl Popper (), probably the most important scientific theorist of the 20th century, the spin - and the pivotal point is: (a) The pre-Popperian inductive-empirical conception. (b) Popper's deductive-falsificationist conception. (c) Post-Popperian conceptions that arose mainly from the criticism of Popper's view.The following examines how scientifically oriented research in psychology is understood under the individual conceptions. 2.1 The inductive-empirical conception of scientific activity This conception of science goes back to the English philosopher Francis Bacon () (Other important representatives are David Hume () and John Stuart Mill ()). In the 20th century, this position was represented in a modern version by the members of the Vienna Circle. Before we turn to the central postulates of this approach, a distinction must be introduced between different types of conclusion Excursus: Different types of conclusions Three basic types of conclusions can be distinguished: Concept 2-1: Types of conclusions 1. Deductive conclusions (Deduction ): These are conclusions that are strictly truth-preserving, ie whenever the premises (= the statements that the conclusion takes as given) are true, then the conclusion (the sentence that is inferred) is also true. Conversely, if there is a deductive conclusion, at least one of the premises must be wrong in the case of a wrong conclusion. The correctness of deductive conclusions results purely from the form of the conclusion and not from the content of the statements.

14 Chapter 2: Conceptions of Psychology as a Science 10 The conclusion of a deductive conclusion cannot lead to any new content that is not already implicitly present in the premises. 2. Inductive conclusions (induction): In contrast to deduction, inductive inferences serve to infer new things on the basis of the given. The prototype of inductive reasoning consists in the development of a legal connection based on available individual observations. 3. Abductive reasoning (abduction): This type of reasoning can be viewed as a special case of inductive reasoning. It is also known as diagnostic inference. The goal of abductive reasoning is to find the set of explanations for a given issue that best explains the issue. The classic case of abductive inference is the solving of a crime involving multiple suspects, and of course it is Sherlock Holmes who, through his close observations and the abductive conclusions drawn from them (rather than deductive, as he put it), catches the perpetrator. Here are some examples to illustrate the three different types of reasoning: Ex. 2-1: Correct deductive inference: A classic example of a correct deductive inference is the so-called modus ponens: AB, AB In words: Whenever A is true, then it is B. A is also true. Hence, B must be true. A B and A are the premises and B is the conclusion. Example: A: Switzerland is located in Central Europe. B: Switzerland is in Europe. A B: If Switzerland is in Central Europe, then Switzerland is in Europe.

15 Chapter 2: Concepts of psychology as a science 11 Comment on the notation: The symbols A and B are so-called propositional variables (since they are variables, they are set in italics in the text). They represent some sort of statement that can be either true or false. You can therefore replace A and B with concrete sentences, as in the example above. The result of an expression like A B is, in turn, a sentence that is either true or false. The symbol is a so-called logical connective, also called a junction (German: logical connection). Specifically, it is about the material implication. This is colloquially expressed as "if then". This type of verbal reproduction hides an important aspect, namely that the implication A B is always true if A is false (regardless of whether B is true or false). Accordingly, the statement "If there are people on Mars, the earth is flat" is true (assuming that there are no people on Mars). If, on the other hand, A is true, then B must also be true if the statement A B is to be true. The example also shows that the two statements A and B do not have to be related in terms of content. The truth value of A B results only from the truth values ​​of the statements involved: A B is only false if A is true and B is false, in all other cases A B is true. The symbol represents the instruction that from the premises (= the set of statements before the symbol) the conclusion (the statement after the symbol) may be inferred. A second example for a correct deductive conclusion is the so-called modus tolens: A B, B A In words: Whenever A holds, B does not hold. Hence A cannot be true either. Example: A: Switzerland is located in Central Europe. A: Switzerland is not located in Central Europe. B: Switzerland is in Europe. B: Switzerland is not in Europe. A B: If Switzerland is in Central Europe, then Switzerland is in Europe.

16 Chapter 2: Conceptions of psychology as a science 12 Comment on the notation: The symbol represents the negation: Whenever a statement A is true, A is false and vice versa. Ex. 2-2: Incorrect deductive conclusions: Two typical examples of incorrect conclusions are: 1. The negation of the antecedent: AB, AB 2. The affirmation of the consequence: AB, BA The following distinction is a central component of scientific language and should therefore be understood. Concept 2-2: Necessary and sufficient condition Given: The sentence A B (where A and B represent concrete sentences). The condition specified by A is a sufficient condition for the condition specified by the sentence B (in short: A is a sufficient condition for B). The condition specified by B is a necessary condition for the condition specified by the sentence A (in short: B is a necessary condition for A). A is sufficient for B, since if A is present (i.e. A is true) B must also be present (according to modus ponens). B is necessary for A, because if B is absent, A must also be absent (according to modus tolens). Ex. 2-3: Necessary and sufficient condition: Consider the sentence: If student S has a master’s degree in psychology, she also has a bachelor’s degree in psychology. The statement can be interpreted as a conditional sentence: A B, with the sub-clauses: A: S has a master's degree in psychology. B: S has a bachelor's degree in psychology. The following applies: The Master in Psychology is a sufficient condition for the Bachelor in Psychology, while the latter is a necessary condition for the Master in Psychology.

17 Chapter 2: Concepts of psychology as a science 13 Comment on the meaning of the logical connections: The meaning of the logical connections can be represented with the help of so-called truth tables. For each combination of the two truth values ​​(true or false), these give the truth value that results from using the juncture for the resulting statement (see Tab. 2-2 on page 68). Comment on the objective of logic: It may seem strange that the implication A B is always true when the statement A is false. The reason for this is as follows: Logic is not concerned with the truth of statements, but with the formal validity of conclusions. The central question is whether the relationship between the premises and the conclusion is such that the conclusion necessarily follows from the premises. This means that a false conclusion can never follow from true premises. Such conclusions must be strictly ruled out. Conversely, a true or a false statement can follow from false premises. Before we finally leave the field of deductive inference and turn to the other two types of inference, I would like to make a brief note on the importance of deductive inference. Note on the importance of deductive inferences: At first glance, deductive inferences seem to be rather useless, since as outlined above in Concept 2-1 the conclusion does not contain anything that is not already implicit in the premises. This impression is deceptive, because in many cases the number of premises is high and it is unclear what the consequences of these are. A formal derivation can show that a conclusion actually follows from the premises. Such problems can be found e.g. in legislation, where it happens again and again that a new law in combination with others has undesirable consequences, which are not noticed immediately (one example concerns the creation of undesirable "tax loopholes" in tax legislation). We now turn to inductive inference.

18 Chapter 2: Concepts of Psychology as a Science 14 Ex. 2-4: Inductive Inferences: Here is a list of examples of the most important forms of inductive inferences. 1. Conclusions from individual observations on a regularity: All of the swans I observed were white. All swans are white. 2. Statistical generalizations: In a sample taken at the University of Friborg (N = 100), 90% of the female students were unmarried. 90% of the female students in Friborg are unmarried. 3. Conclusions by analogy: An experiment with rats showed that the substance X456 caused cancer from an amount of 100 mg. The substance also causes cancer in humans. 4. Causal conclusions: When people shop with credit cards, they pay less attention to costs. Credit cards tempt you to spend money. Here are two examples of abductive conclusions: Ex. 2-5: Abductive conclusions I: The people who saw the film were "invited" afterwards. Possible explanations (causes): 1. The violence depicted in the film increased the aggression of the audience. 2. The exploitation of individuals by state organs shown in the film made people aggressive. 3. The shameless enrichment of some of the people portrayed led to increased aggression among the audience. Abductive conclusion: the last explanation is the most plausible. Example 2-6: Abductive conclusions II: Hans gave Max a "push" at the party. Possible explanations (causes): 1st person: Hans is a person who is easily prone to violence. 2nd object: Max had provoked Hans.

19 Chapter 2: Concepts of Psychology as Science Situation: There were too many people in too small a space at the party. This "crowding" led to increased aggressiveness among the guests. Abductive Inference: The last explanation is the most appropriate. Comment: At first glance, there does not seem to be a clearly discernible difference between causal conclusions (cf. ex. 2-4) and abductive conclusions. However, this is not the case. In the case of causal inferences, it is about the development of a causal connection between a possible cause (use of a credit card) and the effect (increased spending of money). The focus of interest here is the problem of whether there is a causal relationship at all and, if so, via which causal mechanism the cause leads to an effect. In the case of abductive conclusions, the situation is completely different. This is not about discovering new causes or causal mechanisms. Rather, the focus is on which of the possible causes, which have already been identified as possible candidates, best explains the current events. After this little digression on different kinds of conclusions, we return to our topic. Principles of the empirical-inductive characterization of the sciences The empirical conception of science is based on two closely related principles: Principle 2-1: Central principle of empiricism: Any kind of knowledge about the world arises from experience. In particular, scientific laws and theories result from the application of induction. Laws and theories are generated on the basis of individual observations via inductive conclusions. According to principle 2-1, correctly operating scientists choose the following procedure. They try to observe natural processes as closely as possible and, on the basis of observed regularities, induce scientific laws and theories.

20 Chapter 2: Concepts of psychology as a science 16 Ex.2-7: Application of induction I: Classical conditioning: Pavlov noticed that the dog he was observing secreted saliva as soon as a bell rang, which had previously been the food for a long time Dog had announced. He therefore concluded the following legal connection: If a stimulus CS (bell) is paired often enough with another stimulus US that triggers a certain reaction CR, the presence of the CS alone (without the presentation of US) already leads to CR. Comment on the notation: CS = conditioned stimulus; US = unconditioned stimulus; CR = conditioned response. Example 2-8: Application of induction II: Center for letter recognition in the brain: Based on fmri studies, it was found that a certain region in the left lower temporal lobe is activated when reading (Dehaene, 2010). From this the following relationship is concluded: The region in question represents letters. Example 2-9: Application of induction III: Investigations to discover connections In clinical and applied areas of psychology, empirical investigations are often carried out in which as many variables as possible are measured. The aim is to find possible relationships between the variables, e.g. which variables are related to mental well-being or mental health. Such examinations are often (somewhat derogatory) referred to as "shotgun examinations", since, similar to shooting with shot, the aim is to achieve the widest possible spread by including a large number of variables. These examples make it clear that research in psychology can, at least in part, be understood using Principle 2-1. Reference should now be made to a second principle, which is characteristic of late empiricism and was viewed as part of the so-called "received view" (a summary of the principles of late empiricism can be found in Soup, 1977, pp. 50-52).

21 Chapter 2: Concepts of Psychology as a Science 17 Principle 2-2: Separation of theoretical language and observation language: The following two languages ​​must be strictly separated: (a) The theoretical language contains theoretical terms which do not refer to directly observable entities. (b) The language of observation only contains terms that relate to what is directly observable. Theoretical terms get their meaning from the fact that they can either be reduced directly to observation terms, or indirectly to other theoretical constructs, which in turn can be traced back to observable quantities. Theoretical terms for which this does not apply (such as instinctual energy or energetic body zones) are meaningless. The following example illustrates the distinction between theoretical constructs and observations, as well as their relationship: E.g. 2-10: Facets of the concept of the control belief (according to Krampen, 1991) Given: The theoretical construct of the (subjective) control belief, consisting of the 4 theoretical sub-constructs : 1. Internality-externality: conviction regarding the locus of (failure) success. 2. Dependence on power: assessment of the influence of external power. 3. Random influences: assessment of the importance of chance. 4. Competence: Assessment of one's own competence. In Fig. 2-1 the theoretical constructs are symbolized by ellipses. Each of the 4 sub-constructs is measured on 4 subscales of a questionnaire. These measurements form the concrete observations. They are symbolized in Fig. 2-1 by the rectangles. The small circles marked with i symbolize so-called error terms. They represent influences on the measurements that cannot be explained by the constructs in the model. The sub-constructs of the belief in control are directly related to the observed variables. This is represented by the arrows in Fig. 2-1. These express a causal relationship, i. H. Each sub-concept has a direct influence on the people's answer to the associated questions, which are intended to measure the sub-concept (but not on the others). They therefore have a measurable effect and thus acquire their meaning.

22 Chapter 2: Concepts of psychology as a science 18 1 II 2 I 3 Internality 4 IM 1 M 2 M 3 M 4 Z 1 Z 2 Z 3 Dependence on power General conviction of control Random influences 12 Z 4 13 KK 2 K 3 Competence 16 K 4 Fig 2-1: Control Belief Model. The construct of the general belief in control, on the other hand, has no direct relationship to the specific measurements. However, since this construct is directly connected to the sub-constructs (in that it influences them causally), it receives its meaning from them. The model shown in Fig. 2-1 is a so-called factor analytical model (whereby in the current case the relationships between the various variables were interpreted causally). The method of (exploratory) factor analysis is widely used in psychology. On the one hand, it fits very well with the inductive-empirical conception, but can also be considered one of the most

23 Chapter 2: Concepts of Psychology as Science 19 abused statistical methods. The following little digression sheds light on these aspects in more detail: On the importance and misuse of factor analysis in psychological research The factor analysis exists in two basic varieties: As an exploratory or a confirmative technique.In the following we only consider the exploratory variant (e.g. 2-24 on page 44 demonstrates the use of a confirmative factor analysis). Therefore, when "factor analysis" is mentioned in the rest of the presentation, only the exploratory variant is meant. Exploratory Factor Analysis (EFA) is commonly used for developing psychological tests such as intelligence or personality tests. Method 2-1: Exploratory factor analysis (EFA): The application of the explorative factor analysis (EFA) takes place in three steps: 1. Data collection: As many variables as possible are measured which could be relevant for the current area (note the relationship here with the shotgun examinations discussed in Ex. 2-9). The covariances or correlations between all variables are calculated and combined to form a covariance or correlation matrix. 2. Factor extraction: On the basis of the covariance or correlation matrix of the variables, the smallest possible number of factors are extracted with the aid of a statistical procedure (in the case of the so-called maximum likelihood EFA) or a »pseudostatistical« (such as the very often used main factor analysis) . 3. Finding a simple factor structure: A simple factor structure is sought with the help of a so-called factor rotation. This means that the measured variables only load a few factors (ideally onto one) (cf. the model in Fig. 2-1, where each observed variable is only influenced by a single factor). Variables that do not load any factor are eliminated.

24 Chapter 2: Concepts of Psychology as a Science 20 Comment on the main factor analysis: Above, the main factor analysis, which is the most widely used factor analysis method in psychology, was referred to as the "pseudostatistical" method. The reason for this choice of terminology can be found in the fact that the method does not make any assumptions about distribution. Therefore, there is no statistical test to test how well the model explains the data. In the case of the maximum likelihood EFA, there is a statistical test that provides information about the model quality. The EFA can be viewed as a method which induces theoretical constructs based on concrete observations. It therefore fits in ideally with the inductive-empirical conception of science. At this point, however, the criticism of the misuse of the EFA begins, which can be summarized as follows: The EFA replaces thinking about theoretical (mental) constructs and their interrelations with a mechanical procedure. This leads to theoretically uninteresting and unproductive theories. The following example serves to illustrate this criticism. Example 2-11: Factor analysis and personality research: In the following example, Seymour Epstein (1994) caricatures personality research. He is particularly interested in research on the so-called Big Five (see e.g. McCrae & John, 1990). The main result of this personality theory says that personality is made up of 5 factors, with each factor in turn having numerous sub-factors (facets). The 5 personality factors are: 1. Extraversion 2. Compatibility 3. Conscientiousness 4. Neuroticism 5. Openness to new experiences Here now Epstein's story: Once upon a time there was a psychologist named Sam. He decided to examine cars instead of people. He believed it was more scientific to study cars because they are more controllable and easier to understand than people. Sam pointed out that cars have a lot in common with people, such as exercise, eating (i.e. consuming gasoline), breathing (i.e. consuming oxygen), excretion (i.e. producing exhaust gases), individual personalities, and malfunctions.

25 Chapter 2: Concepts of Psychology as a Science 21 Therefore, the knowledge he gained from studying automobiles should be transferable to people. “One by one,” he said to himself, “as soon as I've learned, the personalities of automobiles I can dare to represent people exactly. «Sam now began to determine the fundamental characteristics of cars. With the help of a factor analysis, he revealed five independent characteristics: color (black, white, red, blue, etc.), type (station wagon, sports coupé, limousine, leisure vehicle, etc.), size (compact, medium, full size), maximum speed and robustness (service statistics). The results were most impressive. Experts could use the classification reliably and it was comprehensive (including all types of cars). The system applied within a culture and also between different cultures and all kinds of interesting observations correlated with it. For example, large families with pets prefer station wagons, while singles with a great need for stimulation tend to prefer sports cars. The solution, however, was not as simple as it seems. Sam had to work hard to finally determine the basic properties, but thanks to high-performance computers he was finally able to complete the task. Here is an example of the difficulties that arose: Sam had previously included other dimensions that he thought were important. The factor analysis taught him, however, that these basic dimensions are worthless because they are not independent. This included scientifically worthless variables such as weight, length, width and height. After the factor analysis treatment, this resulted in a single factor: density. At first this seemed like an elegant solution, but then Sam found that density was not independent of size, and the latter explained more variance than density (in a regression equation). So he had no choice but to drop the variable density. Sam took great pride in the level of organization he had brought into classifying cars with just five factors. One day, while he was on his way to presenting his latest results at a psychological convention, his car broke down. Unfortunately, he hadn't the faintest idea what to do. He didn't know anything about how a car worked. In fact, he had never looked under the hood. He had always equated what was beneath it with the deep, dark unconscious of man and therefore considered it unimportant. So he had to be content with what he had, that is, all kinds of research on how the basic properties of a car are correlated with the occurrence of a breakdown. Color usually has no significant relationship with the breakdown statistics, although there is an insignificant tendency for people with red cars to produce more accidents. Robustness, on the other hand, was a relevant parameter and of course predicted future breakdowns by accident. Now he understood why his car broke down: it had a low impact on the robustness factor. It occurred to him that this could easily be carried over to human behavior by substituting neuroses for mishaps. Now he was finally able to explain why people had psychological symptoms: Because they were neurotic. People

26 Chapter 2: Conceptions of psychology as a science 22 develop symptoms because they have a high value on the neuroticism dimension, just as cars break down because they have little value on the robustness characteristic. Unfortunately, his car did not share Sam's enthusiasm for this new finding and did not move an inch. Ultimately, Sam decided that cars were too complicated after all. So he turned back to the study of people. He maintained the dream that by extracting the individual factors by means of factor analysis and uncovering further facets he would finally be able to determine the characterization of every kind of human condition. This meant complete understanding for him. (Adapted from Epstein, 1994, page; translation by the author). Epstein's criticism boils down to the fact that personality research in the area of ​​the Big Five does not provide any fundamental knowledge about the function of the human personality. The results can at most be viewed as a proto-theory. Note: The fact that the Big Five personality research does not provide fundamental insights into the human personality does not mean that it has no practical use. The benefit lies in the fact that the personality factors can have a prognostic value with regard to important areas of life such as life satisfaction or professional success (see the story of Sam) Operationalism: A strict form of empiricism Empiricism assumes that theoretical constructs are only possible through their (direct and indirect) Connection with observed quantities get their meaning. The meaning of the theoretical constructs is therefore exhausted in the various facets of their manifestations. The meaning of the theoretical construct fluid intelligence therefore consists purely in the various forms in which the concept is empirically noticeable, e.g. through high performance in handling abstract problems. A special form of empiricism is so-called operationalism, which was represented by the two physicists Ernst Mach () and Percy Bridgman (). Concept 2-3: Operationalism According to operationalism, the meaning of a theoretical term consists purely in its operationalization. This means that the meaning of a theoretical term arises purely from the way in which it is measured.