n." .- a... . ......— ....... w- ..... «n u c SITUATIONAL ANXIETY AND BIRTH ORDER AS DETERMINANTS OF DOGMATISM AND AUTHORITARIANISM Thesis for the Degree of Ph. D. MICHIGAN STATE. UNIVERSITY , Jacob Jacoby . 1966 EwT‘gt-M‘ 3);“ “if? r THESIS _,,~__ 1' g, gt {“4 ' ‘W:\::Aii;:1,: . ”it” fill-4.1; U mm; This is to certify that the thesis entitled SITUATIONAL ANXIETY AND BIRTH ORDER AS DETERMINANTS OF DOGMATISM AND AUTHORITARIANISH presented by Jacob J aooby has been accepted towards fulfillment of the requirements for PhoD 0 degree in PS I Chfllog Major professor Date__ ZflVlw/b 3 /766 O-l 69 +" m | I I [Inumnmn'mzmn'mnmnm L 312930085 339° e H0 ‘3’ L\_. #255 ABSTRACT SITUATIONAL ANXIETY AND BIRTH ORDER AS DETERMINANTS OF DOGMATISM AND AUTHORITARIAN ISM by Jacob Jacoby The purpose of this study was to determine the effects of anxiety and ordinal birth position on domatism and authori- tarianism. Two multiple hypotheses were proposed: Emthesis I: Individuals placed in threatening, anxiety evoking situations: (a) will manifest a constriction of their belief systems and this contraction will vary with increasing threat; and (b) will become more authoritarian and this increase in authoritarianism will vary with increasing threat. hypothesis II: As anxiety is introduced into a situation and its level increased, compared to later born individuals, first and only born individuals will become: (a) more dognatic; and (b) more authoritarian. A 3:ch2 factorial study, involving 120 Michigan State Uni- versity undergraduates, was conducted to test the hypotheses. The three factors employed were sex (male vs. female), ordinal birth position (first and only home vs. later home), and treatments (hi-threat, mild-threat, and controls). The subjects in the hi-threat condition were led to believe that they were going to receive painful electric shocks, while the subjects in the l 2 Jacoby mild-threat condition believed they were going to receive mild, pleasurable electric shocks. The subjects were run in a randomized, counterbalanced order. Using two matched Dogmatism and Authoritarianism Scales especially con- structed for this study, each subject's level of dognatism and authori- tarianism was measured prior and subsequent to undergoing the treatment effect. Analyses indicated that the threat manipulations were highly successful (i.e., at statistically significant levels). However, no support was found for any of the hypothesized relationships. Several possible explanations for the failure to confirm the hypo- theses were examined. The most likely explanation entailed a distinction made by some investigators between fear and anxiety. Pre— dicated upon this distinction, procedural modifications were suggested for future tests of the hypotheses. SITUATIONAL ANXIETY AND BIRTH ORDER AS DETERMINANTS OF DOGMATISM AND AUTHORITARIANISM by Jacob Jacdby A Thesis Submitted to , Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1966 ACKNOJIEDGEMENTS I would like to take this opportunity to indicate my appreciation to the U. 3. Public Health Service, National Institute of Mental Health, for granting a pro-Doctoral Fellowship to me. It is doubtful that this study would have proceeded as rapidly, or even at all, had it not been for the research funds made available as part of this fellowship. For their critical and insightful guidance, a debt of gratitude is also hereby acknowledged to the members of my committee: Dr. Santo F. Camilleri, Dr. Donald H. Johnson, Dr. Gerald R. Miller, and Dr. Hilton Rokeach (Chairman). My special indebtedness to Dr. Rokeach for the part he has played in the overall direction of my graduate career is also warmly cited. Sincerest thanks are also extended to Colonel Kenneth H. Powers, USAF, for relieving me from my military duties for two months in order that I might complete the first draft of this manuscript, and to Lieutenant Colonel Bernard J. Wiest, USA, Ph.D., for providing me with a quiet office, a pleasant working atmosphere, and all the facilities necessary to achieve this objective. Lastly, I would like to dedicate this volume to Francine, my wife, for the part she played in bringing this study to fruition. ii Chapter I: Chapter II: TABLE OF CONTENTS Introduction The Relationship of Anxiety and Birth Position to Dognatism and Authoritarianism Anxiety as a Detezminant of Dognatism and Authoritarianism Birth Order as a Determinant of Dogmatism and Authoritarianism Summary of Hypotheses and Puzposes of this Study Chapter III: Methodology A Brief Overview Subjects Randomization Procedures The Autho ritarianism and Do gnatism Measures The Anxiety Indices Experimental Assistants Physical Layout of the Ehcperimantal Situation Procedure Chapter IV: Chapter V: References Analyses and Results Discussion and Conclusions iii .15 31 32 32 32 32 33 39 1:0 I40 1:3 50 .72 .81 Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table 10 11 12 13 1h 15 16 TABLES Correlations between Dogmatism and Anxiety. Correlations between Anthoritarianism and Anxiety. Scale of Origin (i.e., Dogmatism or F) Means, and S.D.'s of the A.and B Form Items. (N.B. - values presented are those obtained by Kerlinger, 196A.) Reliability Coefficients for chmatism and F sub- ecales: Reliability Study. Reliability Study: means and S.D.'s. Dogmatism and F sub—scale Mean Aux-I Score fer the 83 run by the three Assistants. Analysis-of-Varience for Differences in Anx-I Score fer the 83 run by the three Assistants. Anx-I Score Means and S.D.'s for the Relevant Sub-Groups. Analysis-of-Variance of the Aux-I Scores for the Treatment (A), Sea: (a) , and Ordinal Birth Position (C) variables. Analysis-of-Variance of the Difference Between Anx-I and Aux-II Scores Across the Treatment (A), Sex (B), and Ordinal Birth Position (C) Variables. Chi Square Test of Responses to the Terminal Anxiety Question. Reliability Coefficients for Dogmatism and F sub-scales: Dissertation Study Pretest Reliability Study: sub-scale means. Dogmatiam and F Dissertation Study (Control group): and F sub-scale means. Dogmatism Change in Dogmatism (dB) and Change in Authori- tarianism (dF) Scores for all Subjects. Means and S.D.s of the Difference Scores fer Dogmatism. iv 11 36 37 38 51 51 52 53 5h 55 55 57 58 6O 61 Table Table Table Table Table Table 17 18 19 20 21 22 Means and S.D.s of the Difference Scores for Anthoritarianiam. Analysis-ofévariance for the Difference in Dogmatism Scores on the two Forms as a Function of Anxiety Level (A), Sex (B), and Ordinal Birth Position (C). Analysis-of-Variance for the Difference in F-Scores on the two Forms as a Function of Anxiety Level (A), Sex (B), and Ordinal Birth Position (C). Mean Changes in Dogmatism.and F Scores. (Where the 80 experimental subjects are grouped according to their Aux-II Scores.) Mean Increases in Dogmatism Scores for HiqAnx Treatment Subjects. (Where the "Terminator" vs. "Non-Terminator" distinction refers to the individual's response to the terminal anxiety question.) Mean Increases in F-Scores fer Hi-Anx Treatment Subjects. (Where the "Terminator" vs. "Non- Terminator" distinction refers to the individual's response to the terminal anxiety question.) p. 62 63 63 70 71 Figure 1 Figure 2 Figure 3 FIGURES Physical layout of the experimental situation. p. kl Mean Change in Dogmatism scores obtained for first and only borns vs. later borne in each of the three treatment groups. p. 65 Mean Change in Authoritarianism scores obtained for first and only horns vs. later borns in each of the three treatment groups. p. 67 Appendix A: Appendix B: APPENDICES Pannes version of the Rokeach Dogmatism Scale. p. 87 Forms used in the Dissertation Study (including the preliminary demographic data form, F and Dogmatism sub-scales A and B, and the Anx-I, Anx-II, and Terminal Anxiety measures.) p. 95 vii Chapter I: Introduction When considered from a heuristic standpoint, the now classic volume The Autho ritarian Personality (Ado rno , Frenkel-Brunswik, Levinson, and Sanford, 1950) must be regarded as one of the most important contribu- tions to social psychology in the past two decades. Reference to both the theory and the resulting personality inventory, the Authoritarian Personality or F-Scale, are to be found in all current introductory texts which attempt to cover the breadth and scope of social psycholog. The fundamental concept underlying the F—Scale is essentially a very simple one. Namely, that an individual possessing a particular cluster of deep-rooted personality characteristics is predisposed to respond in a prejudicial manner to certain social situations, i.e., at least some social attitudes are a function of certain basic personality traits. A more recently developed scale which also links personality factors to social attitudes is the Rokeach Dognatism Scale (Rokeach, 1956, 1960). As will be described below, the Dogsatism Scale is, theoreti- cally speaking, a more broadly based scale which was developed in response to certain limitations inherent in the Autho ritarian Person- ality Scale. The rationales developed for the Dognetien and F-Scales have several factors in cannon. One such commality involves the way in which developmental factors are believed to be responsible for autho ri- tarianism and domatism. Both Rokeach and Adorno et a1. would agree that certain types of experiences during childhood are more likely than others to culminate in either authoritarianism or dognatism. i.,| u. . it. "u . .,. 2 As will be elaborated below, a primary concern of the present study is to focus upon these developmental factors from a somewhat different per- spective than the one normally adopted in the past. A second major concern involves the effect(s) that different de- grees of situational threat have upon an individual's level of dogmath and authoritarianism. Based primarin upon theory in the case of dogma- tism, and both theory and the results of mpirical studies in the case of authoritarianism, a position is adopted which predicts changes in both the amount of dogmatism and authoritarianism as a function of in- creases in situational threat. It is these two concerns which form the basis for the present study. When appropriately translated, they furnish the hypotheses which guided this research. Consequently, the brief overview presented above will now be more fully elaborated and placed in terms of historical and theoretical perspective. In The Authoritarian Personalifl (Adomo, et al., 1950) the authors present the theory and results of a series of investigations which were conducted over a period of seven years. Originally the focus of these investigations was anti-Semitism, an issue of major concern in 191:3. For these early studies a questionnaire — the Anti-Semitism (1:5) Scale -- was devised which would fairly accurately represent each res- pondent's position with respect to anti-Semitism. Interest then shifted to the study of ethnocentrism, or from a specific prejudice (i.e., anti- Semitism) to more general prejudice involving intolerance of all outgroups. Another scale was deve10ped, the Ethnocentrism (E) Scale, to accurately index each subject's level of general ethnic prejudice. 3 In the course of these investigations, data relating to the under- lying personality structure of the individuals were also collected. On the basis of these data the investigators were able to isolate a cluster of personality characteristics which was positively correlated with high scores on the Ans and E Scales. This personality cluster was labeled the authoritarian personality and, in order to get at it more directly and without mentioning the names of specific minority groups, a third scale was devised. This was the Authoritarian Personality or 'F' (for fascism) Scale. Thus, these researchers were able to relate more or less overt social attitudes to deeper underlying personality traits. Within a short period of time The Authoritarian Personality estab- lished itself as a major landmark in the field of social psychology. The F-Scale has now been employed in literally hundreds of investiga- tions (on, Christie and Cook, 1956: Titus and Hollander, 1957) and its underlying tenets have usually been supported. Of course, The Authoritarian Personaliifl has also stirred up much controversy and criticism. (he of the most basic criticisms is that, as a function of the theory that guided its research and the measuring instruments in- volved, The 'Authoritarian Personality focused specifically on "right" or fascist authoritarianism and not on the general properties held in connnon by all forms of authoritarianism (Shils, 1951.; Bokeach, 1956, 1960). Rokeach (1960, p.114) has proposed instead that we "pursue a. more theoretical ahistorical analysis of the properties held in common by all forms of authoritarianism regardless of specific ideological, theological, philosophic,’ or scientific content." h The sentiment underlying this proposal led Rokeach, over a period of years, to develop his own theoretical treatment of general autho ri- tarianism (19514, 1956) culminating in The Open and Closed Mind (1960). Presented in this volume is a theoretical description of specifically what is involved, structurally and dynamically, when an individual is characterized as being either "open" or "closed" minded.1 Coupled with this presentation was an instrument - the Rokeach Dognatism Scale -- which has enabled investigators to treat open and closed mindedness empirically. In the six years since publication of W __Pfi_n_d both the theory and the instrument have been utilized by many in- dependent investigators to study a variety of different variables. In the main, these studies have lent considerable support to Rokeach's theoretical treatment of dognatism. I In brief and simplified version, the conceptual foundation upon which the theory rests may be presented as follows. According to Rokeach, all individuals have at least five different kinds of beliefs about the world which they continuously come into contact with and experience. These five kinds of beliefs form a system in which, in terms of the individual's cognitive functioning, some beliefs are more important than others. In decreasing order of importance, they are as follows: primitive beliefs of 100% consensus; primitive beliefs of 0% consensus; authority beliefs; peripheral beliefs; and, inconsequential beliefs. 1. In the treatment below, general (as opposed to "right") authoritarianism , closed mindedness, and being high in dognatism are phrases used synonymously. 5 Starting from.the assumption that for each and every'belief held by an individual there is (usually implicitly) associated with it a second belief representing that person's estimate of how many other people he feels there are who also hold the first belief, Rokeach de- fines a.primitive belief of 100% consensus as "any belief that virtually everyone [in a position to know J is believed to have also" (1960, p. tr). That is, the individual's second belief is that all those in the total population who are in a position to know about the first be- lief do know, and believe as he does. Thus, if we were to approach anyone who knew me personally and asked him whether he believed ny last name to be Jacohy rather than Smith, we would expect that this individual (and all others like him, i.e., all those to the position of having access to the information) would share my belief about my name. Rokeach defines a primitive belief of 0% consensus as the converse of the above. Instead of virtually everyone serving as external referents or authorities, there is no one. Suppose I suffer from claustrophobia. I have been told many times that my fear is groundless, unrealistic. But it does not help. I go on believing that dreadful things will happen to me in closed rooms. The belief is a primitive one because there exists no external reference persons or authorities who fire in a position to know and whp] can disconfirm it (1960, p. ha). With respect to the third level of beliefs -- the intermediate or authority beliefs - Rokeach reasons as follows. As it is virtually impossible for any one person to know all that he needs to know in order to have meaningful commerce with the rest of the world, much of the individual's information, of necessity, must come from.authorities. The third level of beliefs contains those beliefs one holds concerning the authorities (both positive and negative) that one refers to for this information. The fourth level, called the peripheral level, contains: each and every (nonprimitive) belief and disbelief emanating from positive and negative authority, regardless whether such beliefs are perceived con- sciously as being thus derived by the person himself. For example, favorable or unfavorable beliefs about such things as birth control, the New Deal, and the theory of repression would be considered peripheral beliefs because they are deriveable from the formal content of one 's beliefs abut the Catholic Church, Roosevelt, and Freud. The latter according to our view, would be presented as part of the intermediate region rather than the peripheral region (1960, p. (47). lastly, in the inconsequential region are to be found all those beliefs which are essentially just that -- inconsequential. More specifically, they are inconsequential in teens of the way they relate to and affect the rest of the beliefs in the person's belief-disbelief system. "I would rather take a bath than a shower" might be an example of such a belief. All individuals possess essentially the same belief system dimen - sions. The distinguishing feature of an open system as opposed to a closed system has been described by Rokeach as follows: A basic characteristic that defines the extent to which a person's system is open or closed [$37 the extent to which the person can receive, evaluate, and act on relevant information received from the outside on its own intrinsic merits, unencumbered by irrelevant factors in the situation arising from within the person or from the outside. Ebcamples of irrelevant internal pressures...are...the need to allay anxiety...By irrelevant external pressures we have in mind most particularly the pressures of reward and punishment arising from external authority; for example, as exerted by parents, peers, other authority figures, reference groups, social and institutional norms, and cultural norms. Will the information received about a situation from such external sources be evaluated and acted upon independently or in accord with expectations about how the external source wishes us to evaluate and act upon this information? The more open one's belief system, the more 7 should evaluating and acting on information proceed independently on its own merits, in accord with the inner structural requirements of the situation. ... Conversely, the more closed the belief system, the more difficult it should be to distinguish between information received about the world and information received about the source (1960, pp. 57-58). In other words, the fundamental basis for a distinction between open and closed systms is predicated upon the cogritive functioning of the individual involving specifically his "ability (or inability) to discriminate substantive information from information about the source, and to assess the two separately" (1960, p. 60). The less the ability to do so, the more dognatic or closed minded the individual. The greater the ability, the less dogmatic or more open minded the individual. Chapter II: The Relationship of Anxiety and Birth Position to Dognatism and Authoritarianism 5. Anxiety as a Determinant of Dognatism and Authoritarianism With respect to the etiology of open and closed mindedness, Hokeach has suggested that the "more closed the belief-disbelief system, the more we do conceive it to represent, in its totality, a tightly woven network of cognitive defenses against anxiety" (1960, p. 69). This suggestion was then translated into "the simple hypothesis that those with relatively closed systems should manifest more anxiety than those with relatively open systems" (Rokeach and Kemp, 1960, p. 3&7), and evidence was adduced in its support (see Table l). A study by Fillenbaum and Jaclanan (1961) yielded further corroboration. These authors report finding positive significant relations (cf., Table 1) between scores on the Dognatism Scale and scores on a 39-item anxiety scale developed by Welsh (1956) from the MMPI. Factorial studies by Table 1 Correlations between Do gmatism and Anxiety Study Anxiety measure _I\_T 3; Simificance Rokeach and Kemp (1960) Welsh (1952) Mich. State U. I sample 202 .M. p a .01 New York colleges sample 207 .58 p .... .01 Mich. State U. II sample 153 .61; p s .01 Mich. State U. III sample 186 .57 p .—. .01 English colleges I sample 137 .52 p a .01 English colleges II sample 80 .147 p e. .01 English workers sample 60 .fi p :3 .01 Fillenbwm and Jackman (1961) Welsh (1956) h2 .h9 p.<.01 Rokeach and Fruchter (1956) factor loading 207 .116 chhter, Rokeach and Novak (1958) factor loading 153 .68 9 Rokeach and Fruchter (1955), and Fruchter, Pnkeach, and iiovak (1958) have also indicated the Dogmatism Scale to load highly on an anxiety factor. The relationship between dognatism and anxiety takes on additional meaning when considered in the context of the following passage (Rokeach, Toch, and Rottman, 1960, pp. 376—77): to a large extent the shape of a person's belief-disbelief system is relatively enduring, "carried around" within his personality from one situation to another and accounting for many of the uniformities we can observe in his actions. But this does not mean that‘the situation itself cannot in— fluence a person's behavior. Nor does it mean that a per- son's belief system is open or closed to the same degree at different times. We think of a person's belief system as possessing not only enduring properties, but also more open, Or more closed, in response’to a specific situation in which the person finds himself. We assume that the more threatening a situation is to a person, the more closed his belief system will tend to become. Just as threat or anxiety built into the personality as a result of early experience can lead to closed systems that endure, so should situational thmats lead to similar effects that should last at least as long as the person experiences thmate V Some evidence for this hypothesized relationship between situa— tional threat and dognatism does exist. Assuming that what "holds true of people considemd in isolation should also apply to people who hold beliefs in conmzon and form groups, movements, or institutions" (Rokeach, Toch, and Bottman, 1960, p. 377), these authors analyzed the responses made by the Catholic Church to some of the major crises it has. faced since its inception. The calling of an ecumenical council was operationally considemd to be indicative of some sort of crisis in the Clmrch's history and the canons issuing from these councils (i.e., the actions taken in response to the threat) were independently rated to gauge both the amount of punishment prescribed for violators, IO and the amount of absolute authority implied in it. To ascertain the degree of situational threat, historical accounts of the ecumenical councils were broken down into those events judged to be directly res- ponsible for the convening of the council, and those judged to be merely contributing to it. Two judges independently rated those events on a 7-point scale for degree of situational threat. The reliability co- efficient beween the two judges was .95. Situational threat was then correlated with punitiveness ( r = .52) and absolutism (r = .66). Both correlations were statistically significant, thereby providing empirical support for the hypothesis that as situational threat increases there will be a corresponding increase. in institutional dognatism. However, the relationship between situational threat and (10th has yet to be demonstrated on the individual level. It therefore, seems reasonable to suggest the following hypothesis: Individuals placed in threatening, anxiety evoking situations will manifest a constriction of their belief systms and this contraction will vary with increasing threat. The relationship of anxiety to authoritarianism is also an interesting one. Adorno et al. report that the High Authoritarian, in contrast to the low Authoritarian, tends to perceive the outside world primarily as being threatening (1950, p. 1:11, 1.20). Somewhat in con- trast are the factorial studies by Rokeach and Fruchter (1956). and Fruchter, 'Rokeach, and Novak (1958) which indicate that, as opposed to the Dognatism scale, the F—Scale loads negligibly on an anxiety factor (see Table 2). In line with this finding, Melikian (1956), in a cross- cultural study involving American and mid-Eastern samples, states that "no relationship between authoritarianism and anixety was found in either I “1,! ...Nlit'l! \ x group" (p- 2h?) . Table 2 Correlations between Autho ritarianism and Anxiety 1% Annie measure 1! 3 Simificance Melikian (1956)a Elizur (19h9) 187b .06 N.S. Davids (1955) Taylor (1953) 20 .69 p 4.01 Davids (1956) Taylor (1953) 22 .25 N.S. Singer and Feshbach (1959) Taylor (1953) 1h? .31: p 4 .01 Siegel (19510 Taylor (1953) 99 .25 p < .02° Rokeach and Fruchter (1956) factor loading 207 .27 Fruchter, Rokeach, & Novak (1958) factor loading 153 .38 a. Einployed only 19 of the 29 F—Scale items. b. Ninety subjects were Arabs and 97 were white, angle-saxon, Protestant Americans. 0. As calculated by the present writer. Davids (1955, 1956) reports conflicting results in two attanpts at relating the F-Scale to the Taylor Manifest Annety Scale (W) (Taylor, 1953). In the first study (Davids, 1955) the correlation between the two was found to be +.69 and this was significant at be- yond the .01 level. In the second study (Davids, 1956) a non-signifi- cant correlation cf+.25 was found. Singer and Feshbach (1959) have also studied the relationship be- tween the F and Taylor Scales. In discussing the problem these authors note that: ' 12 Studies dealing with the relationship between authori- tarian tendencies and psychopatholcg have yielded con- flicting rosults...and the issue has elicited considerable ccntroversy....It is rather fruitless to argue what the correlation should be since it is apparently possible to draw diverse Eferences regarding the nature and extent of the relationship, depending upon which statements made by the authors of the Authoritarian Personality... one chooses to stress. Masling (l95h, p. 3115) contends that there is a bias in favor of seeing the equalitarian personality as being "a fairly well adjusted individual" and the authoritarian as "nasty fellow." However, Frenkel-Brunswik (19514) points out that while high scorers may suffer from strong repressive tendencies, the low scorer may engage in ex- cessive self-criticism and introspection as well as manifest other neurotic features. The question, then, of the connection between authori- tarian tendencies and variables usually associated with various states of mental health becomes an empirical issue that can 0 be resolved by the accumulation of pertinent data (p. [401‘ e The pgrungnt dgt‘ pnsmted by Singer and Feshbach indicates a siglifi— cant (p (.01) correlation of +.3h between scores on the two scales. Lastly, based upon a comprehensive trea'hlent of the cultural determinants of authoritarianism, Siegel (195b,) arrived at the following rudimentary hypothesis: Anxiety is a correlate of authoritarianism. Involved in the formulation of this hypothesis is a consideration of the combined worksof Campbell (1952), Centers (19h9, 1952), Davis (1952), Dcllard, £31. (1939), From (l9h1), Horsey (1937), Kardiner (1936), Lynd and mud (1929, 1937), and May (1950). In summary form, Siegel has postulated the following sequence of events: (a) Anxiety and hostility are aroused by the socialization process (Kardiner, From, Ibmey, May, Davis). (b) The mechanism of conformity is employed to allay the anxiety (Kardiner, From, the Iynds, Davis). (c) This conformity leads to a retrenchment into more rigid and conservative economic and social ideologies (the Lynda, Centers, From). l3 (d) The anxiety caused by repressed hostility may cause displacement of this hostility onto minority or socially disapproved groups (Campbell, Dollard, _e__t___a_l.). (e) The more "socialized" the individual, the more amciety may be expected in him, and therefore, the more conformity, rigid- ity, displacement, and in general the more manifestations of authoritarianism may be expected from him (Davis, Kardiner, Fromm). (p. 197). Both F and Taylor Anxiety Scales were administered to 99 female undergraduates at Stanford University. This sample was broken at the median into high and low scorers on the TMAS. The mean difference be- tween the scores of these two groups on the F-Scale was very signifi- cant (t = 3.00, p(.01). Furthermore, the correlation between scores on the TMAS and F-Scale was, r = .25. According to the formula given by Edwards (1960, p. 78), the present writer has calculated this co- efficient to be siglificant at better than the .02 level. Siegel's theoretical analysis and results would appear to posi- tively implicate anxiety as a correlative antecedent of authoritarianism. An examination of the apparent inconsistencies evidenced in the litera- ture (see Table 2) indicates that one of the two studies yielding a non- significant correlation between authoritarianism and anxiety (Davids, 1956) is based on a very small sample, while the other study (Melikan, 1956) utilized the none-too-reliable EILizur (191:9) amciety items and only 19 of the 29 Fchale items. In contrast, the studies reporting significant correlations are, for the most part, based upon sub- stantially larger samples and utilize the more reliable welsh (1952, 1956) and Taylor (1953) Scales. Inspection of Table 2 thus suggests the likely possibility that the "true" correlation between authori- tarianism and anxiety is in the order of .1) and that with repeated sampling of sufficient size, this parameter would be approached. 1h Notwithstanding Singer and Feshbach's (1959, p. 1401;) contention that one is able to draw diverse inferences "depending upon which statements made by the authors of The Authoritarian Personality .. . one chooses to stress ," on the basis of Siegel's (19514) extension it would sew that anxiety is directly linked in a causative way to authoritar- ianism. Thus we may hypothesize that individuals placed in threatening, anxiety evoking situations will become more authoritarian and this increase in authoritarianism will vary with increasing threat. The strength of the hypothesized relationship, however, will probably be less for authoritarianism than for dogmatism. The basis for this assertion is that whereas anxiety is theorized to be the primary causative agent of dognatism, it is only one of several factors implicated in the etiology of authoritarianism. Support for the assertion comes from observing that the average correlation between anxiety and mthoritarianism hovers around .1) (see Table 2) while the corresponding average correlation obtained between anxiety and dogm- tism is in the order of .50 (See Table 1). Generally speaking, amciety thus appears to explain only 9% of the authoritarianism variance while explaining 25% of the dognatism variance. Furthermore, as Rokeach and Fruchter (1955; as well as Fruchter, 2131., 1958) have pointed out, although the F and Dognatism Scales are highly intercorrolated (r: .614) , F is relatively independent of anxiety while Dognatism is not. 15 13, Birth Order as a Determinant of Dogatism and Authoritarianism In seeking to explicate the precise nature of the relationship be- tween dognatism and anxiety, Rokeach and Kemp (1960) provide data "which suggest at least partly that threat and its effects on the closing up of belief systems has its origin in childhood emeriences" (p. 103). These childhood experiences are assumed to be similar to those described by Frenkel~Bnmswik (l9h8) and more fully described in The Auwthorjitarian Personalig. Essentially, the research conducted on these determinants involved extensive interviews of female subjects who scored at the extremes on the F-Scale. An important part of the interview schedule focussed upon the individual's childhood role re- lationships with her parents. Among other things, High Authoritarian subjects typically tended to overtly glorify their parents and to mani- fest greater submission to their parents than did the Low Authoritarian subjects. Futhermore, High Authoritarians manifested a qualitatively different sort of dependence upon their parents than did Iow Authori- tarians. Adorno, et al., described this difference as follows: not being self-reliant, they need support and comfort, first from the parents and then from parent-substitutes. This dependence, however, is neither focussed nor con- scious; it is rather a need for help of others in getting things; the person from when things may be gotten may equally well be parents, or the "leader," or anyone else who seems capable of offering tangible support. The kind of dependence on the parents expected to be characteristic of unprejudiced subjects, on the other hand, is the kind of dependence which people with an ability to love direct to- ward those for whom an object catheads has been extablished (1950, Pa 353). The interview data presented in The .gu'thoritarian Personality is predicated upon what may be termed a cross-familial approach. That is 16 to say, these authors isolated two different modal patterns of child— hood milieus and experiences, and the High Authoritarians were typically found to have had one pattern of familial role relationships whereas the Low Authoritarians usually had the other. However, in light of the plethora of recent empirical studies sparked by Schachter's work on n-Affiliation (1959) and involving ordinal'birth position, it seems reasonable to speculate that childhood milieus might also manifest consistent patterns of igtgg-familial variation. That is to say, all firstAborn children might develop certain role relationships with their parents which are not developed in the cases of second and laterborn children. As an apropos example, Stone and Church (1957) report that: Parental resistance to the child's growing up may stem from unwillingness to relinquisheauthority built up over a decade and a half. We should point out that parental resistance is Often greatest against the oldest child, who in many areas has to break trail for his younger siblings (p. 278). In the volume by Schachter (1959) alluded to above it was found that first and onlyéborn individuals were more anxious in.stress situations and, with anxiety held constant, more dependent upon other persons as "sources of approval, support, help, and reference" (p. 82). (witness the similarity in the type of dependence behavior described here as being characteristic of first and onlyAborns and that noted above by Adorno, et al., as being characteristic of High Authori- tarians.) Schachter attributed the first or onlyAborn's need to affiliate to "child rearing practices as related to ordinal.positinn and of the different consequences of having older or'younger siblings around" (p. 79). He concludes as follows: 17 Influencibility, which is assumed to be in part a function of dependence, is demonstrated to be related to ordinal position. It is anticipated that other dependency-linked behaviors will eventually prove to be related to ordinal position (p. 89). Becker, Lerner, and Carrol (1961;) provide the following inter- pretation of Schachter's findings: It makes sense to suggest as Schachter does, that the more anxious first-born child may have learned to seek out others for support; but it is also reasonable to assume that the later-born child may have had more experience in turning to others, especially peers, as reliable sources of information about the environment. This line of reasoning follows from the fact that the later-born child, having had the presence of an older child as a major agent of his socialization, has characteristically relied upon a comparative peer as a source of validation concerning his beliefs and ideas ... Apparently the older child, through direct teaching or example, charac-r teristically provides the younger child with the "facts of life" and techniques of how to make one's way in the world. The first-born child, on the other hand, does not have this relative peer as a model or source of information (pp. 318-19). In other words, later born children may have a relatively peer oriented model of belief evaluation and validation whereas first and only-hem children may have a more authority oriented model (cf., Warren, 1966, p. 39). Or, as Stotland and Walsh (1963) have aptly put it, as adults "the later home may react as if they were still in a family of peers, which was their initial experience in life" (p. 611:). Becker, et a1. (1961:), continue as follows: If this analysis is valid, and if it is also correct ‘1\ that these early patterns tend to be continued in later life, then it should be possible to demonstrate differ- ) ences bebveen first-born and later-born persons in terms 7 of the way in which they respond to social pressures in K group situations. Because of his greater dependence upon others for emotional support, the first-born person should be more amenable to "normative" influence (Deutsch 8:. Gerard, 1955). That is, he should have a 3 greater need to meet the expectations of other people. 18 On the other hand, the later-born persons should be equally or more amenable to social influence if the content of the suggestion is apparently worthwhile information about the environment. It is our hypothesis that the first- or later-born person might appear to be more or less "dependent" as a function of which type of influence, normative or informational, is operating in a social situation (p. 319). Observe how nearly identical this description is to Rokeach's dis- tinction, presented earlier, of the fundamental difference between open and closed minded individuals. Little meaning would appear to be lost if one substituted open minded for the psychological processes and states characterizing the later born, and closed minded for the psycho- logical processes and states characterizing the first born in either of the descriptions. Rokeach's distinction has been amply supported by both his own research and that of others (e.g., Rokeach, 1960; Powell, 1962; Vidulich and Kaiman, 1961). Becker, et al.'s description has been supported by their own investigations (1961:, 1966) as well as by a recent study conducted by Schachter (1961:) on the sociometric choice patterns manifested by first and later horns. To a significant extent, .4' l Schachter found that first borne tended to establish their friendship 5A _ relationships on the relatively trivial basis of what other peOple thought of the "friend" they had selected rather than upon more rele- vant criteria such as those offered by their own personal needs and interests. Additional evidence which may be interpreted as supporting Becker et al's. (19614) formulations is forthcoming from studies by Dittes (1961) and Radloff (1961). In Dittes' study, later born subjects were found to be virtually unaffected by variations in the degree of peer acceptance, whereas first borne proved to be very sensitive to such 19 manipulations. First borne were more likely to show "blind conformity" and "impulsive judgnents" when they experienced only moderate or less- than-highest peer acceptance. Radloff (1961) reports finding that when uncertain about an opinion for which there mats no objective criteria by which to evaluate its correctness , first borne seek affiliation with others in order to evaluate such opinions via social comparison whereas later horns do not. Apparently than, it seems reasonable to speculate that one possible developmental determiner of the openness or closedness of an in- dividual's belief system might be his birth position. As a preliminary and tentative statement, the following hypothesis may thus be articu- lated: First and only-born subjects will tend to be relatively closed- minded whereas later born subjects will tend to be relatively open- minded. A ‘ ’m And what about the possible existence of a direct relationship be- tween birth order and authoritarianism? In concluding their section on "family patterns", Adorno et a1. (1950, p. 381:) remark that their data "give evidence that the presence or absence of extreme ethnic pre- judice in individuals of our culture tends to be related to a complex network of attitudes within, and relating to, the family". We have already described this research in some detail and noted that its approach was cross-familial rather than intra-familial. In light of the results of the recent investigations by Schachter and others, it would also appear reasonable to hypothesize\t—h—a‘t first and only borne will tend to be, relatively speaking, more authoritarian than later boms e 20 The evidence currently available, however, tends to disconfirm the hypothesized existence of a direct relationship between birth order and either domatism or authoritarianism. A pilot study by the present author furnishes us with the only mpirical data concerning the hypo- thesized relationship between birth order and dognatien. In this study, the Pannee (1962) version of the Rokeach Dognatism Scale (see Appendix A) was administered on an individual class basis to seven (7) sophomore and seven (7) Junior Ehglieh classes at Mason High School (Mason, Michigan) during the week of March 22, 1965. The Pannee version was specifically developed for use with junior and senior high school students. It consists of 15 itws taken verbatim from the Rokeach Dognatiem Scale Form E, and 25 items with modified wording but identical meaning to the remaining 25 items of Form B. The Modi- fied wording resulted from discussions with, and pro-testing on samples in the Junior and senior high school age group. After extensive analysis of these pro-tests, Pannee (p. llo)reporte that based "upon the internal consistency, stability, and discriminatory ability of the instrument to reveal differences in degree of dognatiem, it may be said that Domatiem Scale 'E for Junior-Senior High School is reliable." In addition, Pannee cites several reasons for believing that the validity of her version is also high. Among the reasons given are the following: (a) The items in the revised form have face validity. (b) "Rokeach has stated that the children - revised statements express exactly the same meaning as the original statmnts gathered or ori- ginated by him, used in the original Dogmatiem Scale E" (p. 118). (c) Item ”analysis showed that thirty-nine of the forty statements were 21 very simificant in their ability to discriminate between high and low dognatic responses; the phi coefficient test revealed that thirty-four of the items proved sigrificant at the .05 level or better in ability to discriminate between high and low dognatic responses; tetrachoric r values showed that thirty-six of the forty items indicated a significant relationship between the individual score and the total score. These tests contribute to validity". (p. 118). Three subjects (Se) from the Mason High School sample were elimin- ated because their teachers had identified them as being incapable of reading and comprehending even the revised Pannee items. Eight more forms were excluded from the analysis because the Ss had neglected to respond to some of the items. The data presented are based on the re- maining 370 high school students. The mean score obtained by these subjects on the Pannee version of the Domatiem Scale was 168.28h and the S.D. was 25.251. A Pearson-Product Moment correlation between Dognatism Scale score and birth rank yielded a non-significant 3 of .0887. This is in the opposite direction to that pmdicted as it in- dicates a tendency for the later (rather than the first} home to be higher in Dogaatism. With respect to the mothesized relationship between ordinal position and authoritarianism, a search of the literature reveals only two, studies in which both variables were employed concurrently. In the only reported test for a possible relationship between the two, Greenberg, et al. (1963), gave a battery of tests to 26h Fairleigh Dickinson University evening students. The subjects ranged in age from 18 to 62 and were tested in groups ranging in size from 18 to 1&0. 22 No breakdown according to sex was provided. The subjects were broken down into four groups - only, youngest, middle and oldest -- and the authors reported that the "F-Scale el'nwed virtually no difference be- tween the four groups". (p. 227). The second study (Stotland and Dunn, 1962) provides only indirect evidence concerning the postulated relationship. The sample was com- posed of 298 freshmen and sophomore volunteers at the University of Washington, 10% males and 57% females. In the first emerimental session the subjects were assembled in groups of 70 to 80 and were administered the F-Scale. In the second session, groups of 20 to 30 same-sexed subjects were exposed to experimental. manipulations designed to determine the extent to which each subject identified with c hypo- thetical model individual presented by the experimenters. These investigators report that Low Authoritarians tended to identify with the model whereas High Authoritarians did not (thus confirming other results obtained previously by Stotland and his associates). Further- more, a breakdown of the subjects according to ordinal birth position revealed that the later borne also tended to identify with the model whereas the first and only borne did not. Although no direct analytical comparison was made between ordinal birth position and authoritarianism, the parallel trends indicating that both Low Authoritariane and later borne responded similarly are suggestive. It should also be noted that Stotland and Dunn interpret the trend evidenced in their monograph to be somewhat at odds with Schachter's finding that first and onlies were more dependent upon others than were later borne. They attribute this difference to "be an outcome of the fact that Schachter was dealing 23 with highly anxiety amusing situations, while the conditions of the present simdy were relatively benign" (p. 20). We might imply from the Stotland and Dunn results that ordinal birth position findings will fluctuate as emotional factors, especially anxiety, enter into the experimental situation. This implication seems to find support in the literature. Staples and Walters (1961), using 614 University of Toronto undergraduates in an autokinetic situation, report finding that first and only borne were more suggestible than later borne under anxiety conditions (i.e., threat of electric shock), but manifested virtually no difference under control conditions of no anxiety. However Moore (1964), using 80 male undergraduates in the autokinetic situation, all under non-anxiety inducing conditions, found no sigrificant birth order effects related to suggestibility. Schachter's original. (1959) studies indicated that in the absence of anxiety, first and later borne manifest virtually no difference in their respective desires to affiliate. However, under conditions of anxiety first horns are significantly more desirous of affiliation than are later borne. Sarnoff and Zimbardo (1961) report essentially the same results using a sample of males. Wrightsman (1960) reports that "waiting to gether" as compared to "waiting alone " effectively reduced anxiety for first borne but made no appreciable difference with later borne. Weller (1962), however, in an attempted exact replication of Schachter'e studies, found no birth order effect among 23h females due to the experimental manipulation of anxiety. Other emotional states directly associated with annety have also been manipulated producing differential behavior on the part of first 2t and later borne. Radloff (1961) varied the need for social evaluation and found that "the results of the present experiment complement Schachter's (1959) findings that early and later borne do not differ in overall affiliate tendencies; they only differ when dependency-related ' needs are amused" (p. 583). One can assume that the arousal of such §,_ needs induces a certain amount of anxiety in the first borne. Dittes a. and Capra (1962) report that first borne who are made to feel uncertain about their reaction(s) to emotion arousing threat situations demon- strate greater affiliative tendencies than do other firstborns who are ,, informed of the similarity of their reactions to others. This result was reversed for later born subjects. Dittes (1961) also reports that "qurerimental manipulation of acceptance by peers produced highly signi- ficant differences among firstborn subjects in counter-rejection of group, conformity to norms of group, and impulsive closure on cognitive tasks. Later born subjects appeared invulnerable to differences in acceptance" (p. 358). The association between "feeling uncertain about reaction(s) to threat" (Dittes and Capra, 1962), or "experiences of acceptance and rejection by peers" (Dittes, 1961), and anxiety is readily apparent. Lastly, Gerard and Rabbis (1961) employed 29 male and 71 female undergraduates at Brooklyn College in an experiment involving affili- ation and social comparison. The levels of fear (i.e., threat of strong or weak shock) were employed and the findings, across all birth posi- tions, indicated that the higher the fear the greater the desire to affiliate. When the data were analyzed across both fear treatments, a sigrificant difference was revealed supporting Schachter's finding 25 that first and onlies have greater affiliative tendencies than later horns. When the data were examined separately for males and females, it was found that first and only born women had higher affiliative scores than later born women (thereby supporting Schachter, 1959), but that first and only born males had lower affiliative scores than later born males (thereby contradicting Sarnoff and Zimbardo, 1961). This interesting sex reversal will be more closely examined below. Generally speaking, although some negative evidence exists (viz., Weller, 1962), there does appear to be some degree of consistency in the ordinal birth position literature when emotional states are involved. Especially in those studies which employ anxiety as the independent variable, differences between first and later borne on the dependent variable typically will not be manifested under control (i.e., no anxiety) conditions but will be manifested under experimental (i.e., anxiety) conditions. Of course, it might be argued that first and later borns differ on the level of general anxiety that they bring to the experimental situation. However, Sampson (1965), in reviewing this aspect of the literature, concludes as follows: If we put together these data...two studies indicate that the firstborn have higher anxiety than the later (Schachter, 1959; Yaryan and Festinger, 1961); three in- dicate that the second born have higher anxiety than the first (Rosenfeld, Unpubq Sampson and Hancock, Unpubq Weller, 1962) 3 and two indicate no difference (Moore, 1961;; Weller, 1962). On this basis, it appears that there is no clear conclusion one may reach about the relationship betvwen ordinal position and [initial level of genera1:7 anxiety (p. 196) It is interesting to note that in addition to the Gerard and Rabbis (1961) study (which found that first born women were more affiliative than later born women, but first born males were less affiliative than 26 . // than later born males), other investigators employing subjects of both/ sexes have also reported interesting interaction effects. Sampson \ (1962), for example, has reported that first born males conform more k than later born males, but first born females conform less than later born females. These results were repeated in another study by Sampson and Hancock (Unpub.). Schooler (19614) reports that among female schizophrenics admitted to a mental institution in Maryland during a seven year period there were significantly more last borns than first horns. However, no overall relationship between birth order and hos- pitalization was found among the male patients. Lastly, as was noted above, while Staples and Walters (1962) found (anxious) female subjects to be more suggestible in the autokinetic phenomenon situation, Moore (19611.), also employing the autokinetic situation, reports finding no difference in suggestibility between (non-anxious) first and later born males. In attempting to delineate the possible causal factors which might be responsible for these apparently confusing interactions most authors resort to an interpretation involving differences in socio-cultural child-rearing practices.2 Thus Sampson (1962) suggests that the role “-0-. 2. Relevant here, in terms of cultural differences, is a study by Becker and Carrol (1962). These authors report finding n-Affiliation and. conformity to be related to birth rank in a sample of American boys but not in a sample of Puerto Rican boys. Their explanation is as follows: "If, as Schachter (1959) assumes, the relation between birth order and need Affiliation is due to conditioning procedures specific- ally related to being the first or only born child in a family group, and since child rearing practices are not common to all cultures, it is not surprising to find such a relationship lacking in children of another culture" (p. 131). However, Varela (19614), employing a sample of 66 male and female Uru ayan high school students, does re ort creas- cultural confirmation $.<.10) of Capra and Dittes' (1962 (cont'd.) 27 training given to the female child in our cutlure is closer to the adult role she will play later than is the role training given to the male child, particularly in the early adoption of responsible and independ- ent behavior. A somewhat contradictory interpretation has been offered by Gerard and Rabbie (1961, p. 592): The parent conveys society's norms to the child.‘ One set of norms that is communicated is how a male or f e should behave. In interacting with her male child the mother treats him as a male, rewarding male-like behavior to the exclusion of female-like behavior. The reverse is true for females. The behavior of first born and only children in our emeriment exmplified the Victorian norms of the stalwart male and dependent woman which still probably characterize the ideal man and woman in contemporary western culture. Schooler (1961:) reports one unsuccessful attempt at specifying. the causal factors by an analysis involving information about the behavior of working and middle-class parents towards children of different ranks and sexes. Perhaps, then, in some as yet unspecified way, the sex of the subject also operates as an important mediator of behavior as a function of ordinal birth position. It would thus seem to make good methodological sense for studies involving birth rank to treat the male and female sub-populations separately. In concluding his exhaustive (as of mid-19614) review of the litera- ture, Sampson (1965) has drawn a composite portrait of the first vs. the later born individual. The picture that emerges is the following: 6 The first born child's initial experiences in life are of being alone in an awesome and confusing world. To contend with this world (cont'd) finding that first borns volunteer in greater numbers for small group experiments than later borns (as predicted from Schachter's, 1959, findingS)e 28 he turns to the most readily available source, his parents, and uses them as his model for coping with the complexities that he daily en- counters. However, in very many ways, his parents loom large and dis- tent. The second and subsequent children are born into a much different \K‘ physical and social milieu. Compared to the first born, the second and ‘ j J later born children find a model which is closer, more manipulable, less it powerful, to use in grasping the complexities of their own worlds. Thus: The second child grows up loolcing outward upon a world of peers and learns those skills required for ceping with similars. The first child grows up looking inward, for without there lies a world of still powerful adults, a . more difficult breed to handle, a breed requiring a differ- ent set of skills. (p. 221) However, in certain situations, what appears like a reversal of roles seems to emerge. The inner oriented firstborn turns outward to seek union and agreement with others when his world becomes difficult to handle or issues of choice arise. The outer oriented second born turns inward to seek isolation within himself when difficulties and decisions arise. The power and distance of his parents not only give the first a reduced sense of personal autonomy, but also direct him more to- ward others as useful figures for providing structure, setting direction, and handling problems. On the other hand, the closer model which exists for the second not only permits him to develop a stronger sense of self- confidence, but also instructs him in the more autonomous manipulation of others 3 these turn him back upon his own skills when problems and issues of choice arise. (pp.221-222) Note also a similar interpretation suggested by Stotland and Walsh (1963), namely, that for anxious first horns "the reaction of turning to others may be a socially oriented defense against anxiety, while identification, empathizing, and sympathy @.e., socially oriented 29 behaviorj may be the typical reaction of later horns when they are not personally threatened" (p. 6114). On the basis of Sampson's evaluation and interpretation of the available data, it would appear that certain behavioral characteristics (e.g., patterns of reaction to stress) of first and only born children which are developed as a result of early childhood experiences typically remain latent in the adult. It is only when "difficulties and decisions" arise that these characteristics come to the fore. Conson— ant with this interpretation, the two simple hypotheses presented earlier (i.e., relating dognatism and authoritarianism to birth rank) may be combined and reformulated as follows: As anxiety is introduced into a situation and its level increased, compared to later born in- dividuals, first and only born individuals will become more dognatic and more authoritarian. One final point should be noted concerning the relationship postu- lated to obtain between birth rank on the one hand, and dognatism (or au- thoritarianism) on the other. Though we have conceptualized the former as being a determiner of the latter, consideration of the real world suggests that the relationship cannot be too large. If it were it would mean that a sizeable proportion of only and first born individuals would be highly dogmatic (or mthcritarian) while a corresponding seg- ment of the later born population would be very low in dognatism (or authoritarianism), and this does not appear to be likely. Rather, it seems more reasonable to assume that variations in birth rank are responsible for only a small proportion of the dognatism (or authori- tarianism) variance. Other factors, such as personality of the parents, 3) child-rearing practices, etc., are probably more important determiners. The only point we would like to suggest is that, in addition to these other previously indentified variables, ordinal birth position plays some role in the etiolog of dognatism and this will be more readily apparent in the presence rather than in the absence of situational anxiety. 31 Q. Summazx of @otheses and Purposes of this Stggy By way of sunnnary, this investigation is concerned with the following hypotheses: mothesis I: Individuals placed in threatening, anxiety evoking situations: (a) will manifest a constriction of their belief systems and this contraction will vary with increasing threat; and (b) will become more authoritarian and this increase in authoritarianism will vary with increasing threat. Mothesis II: As anxiety is introduced into a situation and its level increased, compared to later bom individuals, first and only bom individuals will become: (a) more dognatic; and (b) more authoritarian. Two additional purposes of the study are the following: (1) To investigate, without any a priori motheses, the potential effect of sex differences upon the relationships hypothesized above. (2) To help resolve an issue in the literature by providing additional data pertaining to the level of general anxiety first and only barns vs. later boms possess upon initial entry into the experimental situation. Chapter III - Methodology A Brief Overview - A 3xe2 factorial desigx was employed to test the hypotheses. There were three treatment levels (one control and two emerimental groups), two levels of sex (male vs. female), and two levels of birth position (first and only boms vs. later home). The subjects (53) were first administered measures of dogmatism and authori- tarianism. The experimental manipulations, in which the two groups of experimental Ss were made differentially anxious, were conducted and all 35 were then tested again for increases in dognatism and authoritarian- ism. Pre and post measures of anxiety were obtained to check on the efficacy of the manipulations . Subjects - One hundred twenty students enrolled in Introductory Psychology at Michigan State University during the sunmer of 1965 served as Se in the experiment. All Ss were motivated to participate for at least two reasons: (a) they received credit towards their final grade for each hour they served as Se in psycholog experiments, and (b) the experimenter personally visited each class at the beginning of the term and asked the students to participate in this experiment rather than others because "it means my Ph.D." These Ss could not be considered "volunteers" in the tmest sense, however, as most of the instructors made a minimum of 3 hours participation in experiments a requisite for the course. The sample was equally divided into groups of males vs. females, and first and only borns vs. later borne. Randomization Procedures - The 33 Se in each of the four combina- tions (i.e., male first and only borne, male later borne, female first and only borne, female later boms) were randomly assigned to one of 32 33 the three treatment conditions (i.e., High Anxiety, Mild Anxiety, Control). This yielded 3x2x2 = 12 cells, with 10 $3 to a cell. Two equivalent shortened forms of the Dognatism and F—Sceles (hereafter referred to as Form A and Form B) were constructed (as described below) and five Se in each cell were randomly pre-assigned to take the A form first and give to take the B fem first. Thus, since the individual Ss were run successively, given knowledge of an S's sex and birth rank all subsequent procedures were determined. For example, supposing student X was the sixth male later born to arrive for the experiment. The flow chart would show that he was to take Form A first and then be exposed to high anxiety manipulations, after which he was to take Form B. The sequence for the seventh male later born, however, would be to take Form B first, be exposed to mild anxiety manipulations, and then take Form A. (These procedures will be described shortly in greater detail.) The Authoritarianism and imam Measures - In late 196h, Kerlinger had both the F and Dognatism Scales administered to a sample of 537 Introductory Psychology students at Michigan State University. The students responded to each item on a seven point scale ranging from +3 (Agree very strongly)to -3 (Disagree very strongly), with the neutral 0 value being omitted. The mean and standard deviation were computed for each item on the two scales and, in general, the values obtained for the Dogmatism items closely correspond to those reported by Rokeach (1956). The Kerlinger data, made available to this author by Rokeach, were used in the present study as the basis for construct- ing two matched F and Dognatism sub-scales. The procedure was as follows. 3h For both the F and Dognatism Scales, individually, the means were ordered according to magaitude and then paired (e.g., the Dogmatism items with the two highest means formed one pair, the next two highest formed the second pair, etc.). Since the final form of the Dognatism Scale (Form E) contains to items, this procedure yielded 20 pairs for the Dogmatism Scale. However, since the final form of the F-Scale (Forms to - AS) contains 29 items, one F-Scale item had to be elimin- ated. Since we hypothesized an increase in score resulting from the introduction of anxiety, it was reasoned that an item allowing for greater increase in score should be preferred over an item allowing for little increase, as the latter might tend to introduce a ceiling effect. Consequently, since "Human nature being what it is, there will always be war and conflict," was the item with the largest mean (7C = 5.03), it was eliminated. Two matched F and Dognatism sub-scales were then formed by assign- ing the larger mean of each pair to one sub—scale, with the smaller mean of each pair being assigred to the other sub-scale. This was done in an ABBA order. By way of explanation, let us denote the first sub- scale by the letter A and the second sub-scale by the letter B. Now consider only the larger means in the following four hypothetical pairs: 5.0 and (4.9; in? and 14.5; 11.2 and lid: 3.9 and 3.8. The larger mean of the first pair (i.e., 5.0) would be assigned to sub-scale A; the larger mean of the second pair (i.e., h.7) to sub-scale B; the larger mean of the third pair (i.e., h.2) to sub-scale B; and the larger mean of the fourth pair (i.e., 3.9) to sub-scale A. (This means that the smaller means in each pair were assigned in a BAAB, or 35 "mirror-image" order). The larger mean of the succeeding (i.e., fifth) pair was assigned as the initial element in a new ABBA pattern. The resulting means for the two 20 item Dogmatism sub-scales were 7h.675 and 714.821. The mean difference between the two sub-scales is thus a negligible .116. The means for the two 1h item F sub—scales were h8.018 and 148.103. The mean difference between those two sub- scales , .085 , is also negligible. Next, the 11: items of one F sub-scale and the 20 items of one Dognatism sub-scale were integrated into one 31: item form (Form A), and the items of the other F and Dognatism sub-scales were integrated into a second 3h item form (Form B). The sequence in which F or Dognatism items appears in each form was randomly determined (see Table-3). The overall means (based on the Kerlinger data) for these two forms were 122.778 and 122.83h. Copies of the two forms will be found in Appendix B together with the other material presented to the 58. To test their reliability, both forms were administered to a sample of 17h n.s.U. students enrolled in Introductory Psychology during the spring quarter of 1965. (N .B. This will later be referred to as the "Reliability Study".) The resulting raw correlations hovered, for the most part, slightly above r = .6 (see Table 14). Corrected for length by the Spearman-Brown formula, they ranged in the order of .7h to 7 .82. The corrected F-Scale reliability coefficients for the tctal sample (r = .782) falls below the mode ( r = .90) for the 1h relia- bility coefficients obtained by Adorno, et a1. (1950, p. 258) for Forms lo and is of the F-Scale. The corrected Dognatism reliability coefficient for the total sample ( r I .753), however, fall nearly at l‘llll‘lll fi Table 3 Scale of Origin (i.e., Dogmatism or F) Means, and S.D.'s of the A and B Form Items. (N.B. - values resented are those obtained by Kerlinger, 196 .) Form A Form B Item Oriana Mean 5 .D . Item Origiga Mean 1 D 5.7%; 1.582 1 D 5.h73 2 F 3.1 1.822 2 F 3.039 3 F 2.h19 1.391 3 D 5.397 h F 3.050 1.hh0 h F 2.296 5 D 1.862 1.050 5 D 3.199 6 F 3.026 1.662 6 D 3.983 7 F .730 1.78h 7 F h.h25 8 D .2lh 1.809 8 D h.o65 9 D 5.101 1.572 9 F 2.769 10 D 3.0h5 1.621 10 D 3.0h5 11 F 2.912 1.766 11 F 3.h39 12 F 3.769 1.758 12 F 2.957 13 F h.026 1.673 13 F 2.676 1h D 3.937 1.795 1h F 3.890 15 D h.0h3 1.675 15 D b.155 16 D h.056 1.558 16 F h.53h 17 F h.551 1.728 17 D 3.799 18 F 3.03h 1.701 18 D 3.875 19 D .559 1.720 19 D 3.553 20 D .119 1.751 20 F h.808 21 D h.538 1.936 21 F 3.287 22 D 2.795 1.78h 22 D 2.136 23 F 2.629 1.hh9 23 D 3.993 2h D 3.h00 1.850 2h D 3.879 25 D 2.829 1.h18 25 0 2.838 26 F b.700 1.715 26 D 2.912 27 D b.700 1.6bh. 27 F‘ 3.0h1 28 F h.216 1.75h 28 F 2.9h8 29 D 3.803 1.515 29 D 3.236 3) D 3.825 1.876 :0 F 3.909 31 D 2.991 1.831 31 D h.598 32 D 3.1h0 1.h09 32 D b.685 33 F 2.903 1.203 33 D 3.5% 3h D 3.901 1.700 3h D 2.h96 a. D = Dogmatism Scale; F 2 F—Scale 37 the mode (r = .76) for the 10 reliability coefficients obtained by Rokeach (1960, p. 90) for Form E of the Dogmatism Scale. 0n the basis of these results (especially for the Dogmatism subscales) it was felt that further modification of the two sets of sub-scales would not appreciably raise the reliabilities. Table is Reliability Coefficients for Dogmatism and F sub-scales: Reliability Study .1} Doggtism E raw _r corrected 5 raw _r_ corrected _r_' Total sample 17h .60 3h85 .75271 .6b1585 .78166 Males 117 .616108 .7625 .6h2006 .78197 Females .59h2h9 .7h5h9 .612069 .75935 83 taking Form A first 86 .639686 .78025 .5917Sh .7h352 gingfigrst 88 . 589169 . 7111148 .688831 . 8157h The means for the two forms, however, were not nearly as equiva- lent as was expected on the basis of the Kerlinger data (see Table 5). Tests for the significance of the difference between means of corre- lated samples (Ferguson, 1959, p. 1110) were conducted for both the F and Dognatism sub-scales. An insignificant 5 of 1.12 was obtained for the difference between the total sainpie mezms on the F sub-scales. However, a significant (p. (.01) t of 3.57 was obtained for the differ- ence between the total sample means on the Dognatism sub-scales. 38 Table 5 Reliability Study: Dogmatism and F sub-scale means and S.D.'s Matism sub-scales IN Form A Form B 3? S.D. 'i S.D. Total sample 17h 72.603h 11.5091 75.3735 11.5969 Males 117 7h.3761 11.2723 76.h786 11.0933 Females 55 68.5273 11.11.19 72.8182 11.9hh5 88 taking Form A first 86 73.2907 11.1618 7h.b.hl9 10.7677 85 taking Form B first 88 71.9318 11.86140 76.28141 12.31171: F sub-scales N hmA hmB Tc S.D. Ti S.D. Total. sample 17h 16.3908 10.5726 117.1897 10.0922 Males 117 1.8.1111 9.31m 118.5556 9.8168 Females 55 142.3th 11.91116 bh.2182 10.2861 35 taking Form A first 86 1.740117 10.h90h h7.2h1.2 9.6783 55 taking Form B first 88 115.6932 10.6656 1.7.1331: 10.5361. A judgement, therefore, had to be made concerning whether or not to proceed with the two Dogmatism sub-scales as they were and adjust the scores just prior to the analyses, or to construct two altogether new nub-scales in an attempt to obtain equivalent mean scores as well as 39 high reliability. Since construction of the present sub-scales was based on data from a sample (Kerlinger, 19611) more than three times larger than that of the preth reliability study ( 537 > 3 x 171.), and since the obtained reliability coefficients were considered to be fairly high considering the small number of items in each subscale, it was de- cided that the best way to handle the discrepancy would be to add in a correction factor in the analyses of the results should this later prove necessary. Anxiety Indices - Three measures were taken of each S's level of anxiety, one prior to and two after the elqierimental manipulations (see Appendix B). The pro and one of the post measures, hereafter re- ferred to as Anx-I and Aux-II, consisted simply of the S's response to the following 6 point question (adapted from Schachter, 1959, p. 31): How nervous or uneasy do you feel about taking part in this experiment? Please answer by checking one of the following alternatives: I feel extremely uneasy I feel very uneasy I feel quite uneasy I feel a little uneasy I feel relatively calm I feel completely calm The second post manipulation measure (hereafter called the "tamin- al anxiety" measure) is somewhat more indirect and consists of the S's response to the following question. Would you like to leave rather than proceed with the experiment: yes no ho Marinental Assistants - Three male undergraduates were hired as experimental assistants. Their primary function was to conduct the in- troductory work with each of the 53 so that when the E began to work with them he was unaware of their respective birth ranks. The assistants received the following training for the job. The nature of the study was described at length and in detail until the E was satisfied that each assistant had a good understanding of what. was going on. In addition, each assistant was instructed on how to conduct himself with, and what to say to each of the 33 (as described directly below). On the following day, each assistant was required, in private, to describe the procedure in as great detail as he could, focussing specifically upon his part in the experiment. Omissions and incor- rections were attended to. In addition, they were given test protocols to score, and their work was closely scrutinized by the E. (Their work was also periodically spot-checked by the E during the course of the experiment proper.) Each assistant ran three Se in the last phase of his training. These 58 were used solely as training 38 for the as- sistants and their results were discarded. The procedure followed was exactly the same as was followed during the experiment, and the E closely supervised each assistant's performance during this phase. PhysicaluI-gyout of the qulerimental Situation - Two adjoining rooms in the basement of the new Psychology Research Building at M.S.U. were utilized for the experiment. As illustrated in Figure l, the only means of entry into the rear room was by walking through the front room. when the door between the two rooms was closed, the rear room was somldp roof .. .mrr ”IIUIIIIUII (It'll (I'll. '1’, -1 '1' m...- -"I. Q student desk l front reon l mmmox‘oocr assistant 's 3 table +— I ‘1 n -, - __ ._ _ - -. - . ..-- ..---..--“.llnuzunaaaanm'mullz , ,, ~ n t bookcase ] I” .. . I I & nonsense \ \ syllable lists S 's table E's desk rear room Vrrmrrflllmnrrl/Irtll’rtrn. tartar- a/ (Izt'rl 'I'llllltllll'l’u I I Illlllllllllllll llllllllllllllll'ltll a n ..l' I. \ u‘leJLl/Jll/ll/ll/l um rum um v rm 0 I l Illlllllllllll l l u I n 7 uluslulln/xnll ’llllllllllllllllll'l) I/Il W I Fig. l Plvsical lmut of the experimental situation. Note: the arrows indicate the direction in which a seated individual would be facing. hl hz The front room contained a table and chair (for the assistant), a large student desk (for the S) , and bookshelves upon which were ar- ranged most of the paper forms needed for the experiment. The as- sistant's table held a flow-chart indicating how each of the successive Se in the four combinations (male vs. female 1. first and only vs. later born dichotomies) were to be run. That is, the flow-chart in- dicated which form they were to receive first ( A. or B) and which of the experimental treatments (Hi-Amt, Mild—Ame, or Control) they were to receive. The rear room contained the E's desk, a table and chair for the 5, another smaller bookcase, and a large metal equipment rack full of electrical wiring and gadgetry. The chair and equipment rack were at opposite ends of the table so that an individual sitting in the chair would be facing the equipment rack which was standing approximately two feet from the other side of the table. The bookcase was standing adja- cent to the rack and it too faced the chair. In addition to containing the remainder of the papers and forms used in the study, it held some more electrical equipment (which had wires leading to the rack), as well as a conspicuously displayed First Aid Kit. On top of the table was a manila folder bearing the label: "Nonsense Syllable Lists: 3 - practice; Ll- - test A; 5 - test B". Inside were three appropriately labeled nonsense syllable lists. The significance of these lists is described below in the "Procedure" section. The equipment rack was employed to induce the anxiety. It had several alternately flashing red lights, a conglomeration of complexly wired connections, an apparatus which generated a continuous random 143 flow of intense blue sparks across a B/h inch gap, and two dead (i.e., non-functioning) finger electrodes which were long enough to be affixed to the fingers of the individual sitting in the chair opposite. The sparks produced a loud staccato crackling noise and after being in operation for 20 or so seconds, produced a smell similar to that of burning ozone. Procedure - Each 5 announced his arrival by knocking at the front door, whereupon the assistant on duty ushered him in and bade him be seated at the student desk in the front room. The door to the second mom, in which the E sat attired in a white lab coat and working at his desk, was closed. After being seated, the S was presented with a single sheet which requested him to indicate his age; sex; date and place of birth; and number, sex, and dates of birth of his siblings. At the bottom of the sheet was the Anx-I question. (Note: Appendix B contains all of the materials arranged in the same order as they were presented to the 33. All the different sets of forms were sequentially numbered and each successive S was given a number and retained it throughout.) When the S finished responding to this first sheet the assistant collected the paper and noted the response to the birth order question. Given this response, the assistant checked his flow-chart to determine which form (A or B) and which treatment (Hi-Anx, I‘fild-Anx, or Control) the S was to receive. The assistant then took the next numbered copy of the appropriate form from the shelves and administered it to the 3. Up to this point all the 53 were treated alike. Once they fin.- ished responding to the first form, however, they began to be treated 14).; differentially. The first difference came in how they were brought in to the E sitting in the second room. As soon as the Control Ss had finished the first form the assistant said: "Excuse me, but I have to leave for a few moments. You can finish the second part of the questionnaire next door.” At this point he took the S and the second form with him, knocked, then opened the door to the rear room and said to the E: "Excuse me, but I have to go make that phone call now. Would you please finish this off for me? Thanks." As the assistant left, the E gave the S the second form and nodded towards the table and chair saying: "Okay, you can sit there and finish the question- naire. Don't let that apparatus bother you; it's being used in a different experiment also being carried out in this room." The E then resumed working at his desk. After the Control Se had finished res- ponding to the second form the E gave them the two post-«anipulation anxiety measure (see Appendix B) saying: "Now would you please answer these last two questions as honestly and sincerely as you can." Upon completion of the first form the So in the two emerimental conditions were told by the assistant that the mainder of the experi- ment would be conducted in the other room and would they please follow him. After opening the door the assistant told the E either to "run this S as a number 2" (i.e., Mild-Aux) or "a number 3 (Hi-Anx)". In this way the E was able to administer the appropriate manipulations while remaining ignorant of the 8's birth rank. All experimented S's were seated at the table and were first asked to answer a question relating to their general state of health (see Appendix B). This was intended to increase the experimental S's level g 1:5 of anxiety, as described below. Throughout (i.e., during the in- structions and all subsequent conversation with the S), the E spoke in more serious and somber tones to the Hi-Anx 33 than he did to the Mildp Am: So. ' These experimental Ss were instructed as follows: Please have a seat. Before we begin, please answer this question concerning your general state of health. No matter what the 6's response was, the E continued as follows: The reason why we ask this question is because this is an experiment involving electrical shock and we would like to eliminate all those people with weak or deficient hearts. Do you have any heart trouble that you know of? Four 53 responded in the affirmative and, consequently, they were dismissed. For all other Ss the E continued: Fine, now I'll explain some of what's going to happen. In this experiment we are interested in the rate at which people are'able to learn new informa- tion under stress. Here, what we mean by "new information" is lists of nonsense syllables. These are lists of three letter words which have no known ' meaning. As you have probably learned in your Psych. class, the reason why we use nonsense syllables is because Just about nobody has had any prior experience with them. This means that we can assume that everyone starts off at the am basal. level and this greatly simplifies our measurement task. Stress, on the other hand, is defined here in terms of electrical shocks. What we're going to do is first give you a practice list of 10 nonsense syllables so that you can get the hang of it. After you have learned this first list, you will be given a test list to learn. At this point we are going to start giving you a series of electrical shocks and we are going to measure the rate at which you learn this second list while you are being shocked. We are then goingto giveyouathir'dlistto learn, andthis time you will not be shocked. The purpose of this last list is for us to get some comparative measurements on you. 146 The purported purpose of the experiment was given added credibility by the manila folder labeled "Nonsense Syllable Lists" which was lying on the table directly in front of the S. Inside the manila folder were three appropriately labeled nonsense syllable lists to quell the curio- sity of am; 5 who might open the folder. None chose to do so. Up to this point the instructions were the same for both groups of experimental So. In the case of the Mild-Aux 83 the E continued: Don't let the word shock'trouble you; I am sure you will enjoy the emergent. The shocks will be very mildand I assure you that what you will feel will in no way be painful. The shocks resemble more of a tickling or'tingling sensation rather than any- thing unpleasant. We will fasten these electrodes to your hand, hook you into the apparatus and, after the machine warms'up, 'start you on the nonsense syllables. Again, I'do want to emphasize the‘fact that most people report the shocks as being more pleasureable than anything else. The preceding critical paragraph defining stress was deleted and the following one substituted for the. Hi-Anx Ss: Now I feel I must be completely honest with you and tell'you‘smactly what you are in for. These shocks will hurt. first people say they are rather painful. As you can guess, in research like this, if we are to learn any- thing'of real value, it is necessary that our shocks be intense. 'What we will do is fasten these electrodes to your hand, hook you into the apparatus and, after it warms up, start you on the nonsense syllables. Again, I do want to be honest with you and tell you that these shocks will be quite painful but, of course, they will do no permanent damage. These two critical paragraphs were adapted from Schachter (1959, pp. 13-11;). It should be noted that threat of electric shock has been consistently observed to be as effective as actual electric shock in inducing physiOIOgical stress reactions. For example, studies by Ax (1953). Deane (1961), Hodges and Spielberger (1966), and J. Schachter h? (1957). have demonstrated that threat of shock is sufficient to pro- duce significant increases in heart rate (HR) and other indices of autonomic nervous system (ANS) arousal. From.this point on, the instructions to the two groups of experi- mental Ss were the same. E continued: Will you please remove any watches or rings which you have on and.place them.on the side of the table. New, which hand do you.write with? (S responds) Fine, may I have your (Opposite) hand. The E then attached the electrodes to the 8's non-dominant hand and.proceeded to plug the apparatus in. Once plugged in, the apparatus flashed, crackled, and marked ominously. It takes the apparatus several minutes to warm up. While‘we are‘waiting*would.you.please answer the items on this questionnaire. The format is the same as the one you'took in there (E motions to front room), but the questions are different. You.will have enough time to finish it'before we start on the nonsense syllables and shocks. After the S had completed the second form the E picked it up, looked at his'watch, handed the S the last sheet containing the final two anxiety measures, and said: Fine, we still have about two minutes remaining before the apparatus is warmed up enough to deliver shocks. ' Would you please answer these last two questions as honestly and sincerely as you can. It was thought that placing these two measures at the very end of the experiment would indicate whether, in the case of the experimental $8, the anxiety induced by the threat of shock had sustained throughout the period during which they were responding to the second form. Con. sistent with such an interpretation is the report by Hodges and Spiel- berger (1966, p. 288) that: he Deane (1961) found that 55 told to expect shock during a long anticipation period responded with HR acceleration prior to the time the shock was expected and maintained this accelerated HR on subsequent trials even though they were never shocked. Thus, it would appear that _t__h__re___at of shock is as effective in inducing ANS” arounsal as the actual presentation of the shock stimulus. When all the Ss in the experiment -- controls as well as experi- mentals -- had answered the last two anxiety questions they were told that the experiment was over. In the case of the experimental $3, the apparatus was disconnected at this point, usually from very surprised looking 53. The E told each S that the experiment was a very complex one in- volving six emerinental and six control groups and that, strictly on the basis of chance, he had been randomly assigned to one of the control groups. The experimental 53 were told that this (i.e., the fact that they were "control? subjects) was the reason why they weren't shocked, and the true control 33 were told that this was the reason they did what must have appeared to than to have been very little actual work for the experimental credit. All 53 were further told that since there were twelve groups in the experiment, if they spoke about what had happened to them with other students who would be participating in the experiment, these people would arrive with pro-conceived notions which would probably be inappropriate for their particular group and this would tend to distort their responses. It was also pointed out to the experimental Ss that they could easily imagine what would happen to the number of peOple who would volunteer or show up for this experiment once the word leaked out about the electric shocks. Lastly, the cause of scientific advancement was made grounds for another appeal. Thus, h9 it was stressed to each S, in several ways, how exceedingly important it was that he not divulge anything concerning the experiment to any- one. At the conclusion of this talk, each S was asked to personally commit himself to remain silent. This was done by the E asking in conclusion: "Do I have you word, then, that you won't discuss amthing of what you saw or what happened here with anyone?" Only after the S had committed himself was he‘given the credit slip, thanked for his participation in the experiment, and shown to the door. All 58 in all conditions appeared to respond with a sincere "yes" to the comittment question. Furthermore, the E and his assistants were on guard to observe if an' of the in-coming Ss seemed to be in any way knowledgeable of what was to transpire. In only one case did it appear as if an S was aware of what was going to happen, and his re- sults were discarded. It should be noted, however, that none of the Ss were asked directly if they had had any foreknowledge of the experi- mental procedure. It should also be added that no direct credibility measures were obtained at the termination of the experiment. That is, the experi- mental 53 were not specifically asked whether they believed they were actually going to be shocked. However, from the surprised looks ob- vious upon most of their faces when the electrodes were moved, it did appear that they were convinced of the E's sincerity. The results of the analyses conducted on the anxiety measures (see Chapter IV below) provides indirect evidence that this was so. Chapter IV: Analyses and Results In total, 1140 Se were run. Nine served as training 83 for the assistants. The results for a tenth S were discarded because he refused to answer two of the itms on the grounds that he felt absolutely and perfectly neutral on the issues involved. Another S was eliminated be- cause he said he "just knew" he wasn't going to be shocked. Four more 88 were eliminated because they indicated that they had either weak hearts or heart related problems. Lastly, five 83 being run under ex- perimental conditions (two from the Mild—Anx and three from the Hi-Anx conditions) becane so anxious and nervous during the instructions phase that they just would not continue. They all teminated at one point or another during the instructions. Before leaving, each of these 20 83 were given the standard talk on the importance of not discussing the experiment with others. The analyses presented below are based on the remaining 120 58. One of the three exper'ituntal assistants (assistant #2) was European-born and raised, and spoke with a very marked accent. This seemed to introduce the possibility that the assistants might have ex- erted differential effects upon the 83. Specifically, it was believed that an assistant with an accent could potentially raise the Ss' general level of initial anxiety. Using the individual Anx—I scores, a simple one-way analysis-of-variance was therefore conducted to test for such differences. Table 6 contains the respective frequencies, means, and S.D.'s for the» three assistants. Simple inspection reveals that the mean Arno-I score was indeed greater for the foreign born assistant, but not appreciably so, as the difference fails by a substantial enough 50 Table 6 Mean Amt-I Score for the 83 run by the three Assistants EM 3 Mean Aux—I Score Si); l to 1.90698 ’ .78115 2 61 2.13115 .56021 3 16 1.87500 .8062 3 margin to reach an acceptable level of significance (F = 1.537; p>.05; (see Table 7). Table 7 Analysis-of-Variance for Differences in Anx-I Score for the 33 run bv the three Assistants Variance Source Sumof Squares _d_f Mean Square _13 Between Groups 1.6 3792: 2 .8189? 1.5 37 32 Within Groups 62.32873 11? .53272 Total 63.96667 119 Next, a 3va2 analysis-of-variance was conducted to determine whether any group of Se began the experiment more anxious than others. This analysis served two purposes: to help resolve the issue in the literature concerning whether first or later horns are typically more anxious to begin with; and, to enable us to use these Anx-I scores as a base-line for evaluating the increase in anxiety (Arm-II minus Anx-I) as a function of the experimental manipulations. with slight depart- ures for each of the relevant sub-groups (see Table 8), the overall 52 Table 8 Ann-I Score Means and S.D.'s for the Relevant Sub-Grows. 9.12.112 2L lies. in Hales 60 2.000 .6106 Females 60 2.0 33 .8143 First and 60 2.017 .676 3 only Borne Later Borne 60 2.017 .7917 Male First and only Borne 3) 2.033 .6119 Female First and only Barns 30 2.000 .7h28 Hale Later Borns I) 1.967 .6119 Female Later Borne 30 2.067 sum Overall 120 2.017 .7332 mean Ann-I score and S.D. for the 120 38 was 2.017 and .7332, respec- tively. The analyses failed to reveal any differences which even approached significance (see Table 9). Two separate analyses were conducted to test the efficacy of our experimental manipulations. The first of these analyses involved the difference between anxiety scores (i.e., Anx-II minus Amt-I). It was predicted that the manipulations would produce no significant differences between Anx-I and Arno-II across the sex and 0B? variables, but would produce significant differences across the treatment (i.e., anxiety) condition. A 31:22:? analysis-of-variance utilizing the differ- ence scores reveal these predictions to be overwhelmingly continued 53 Table 9 Analysis-of-Variance of the Anx-I Scores for the Treatment (A), Sex (B), and Ordinal Brith Position (C) Variables. Variance Source Sum of Squares _d__f Mean Square _F_‘ Sigiiicance A .2166? 2 .10833 .189 32 N.S. B .0 3333 1 .03333 .0582; N.S. c .00000 1 .00000 .00000 N.S. AB 1.51667 2 .75833 1.32521; N.S. A0 .15000 2 .07500 .1310? N.S. BC .13333 1 .13333 .23301 N.S. ABC .1166? 2 .058 33 .1019h N.S. Error 61.80000 108 .57222 Total 63.96667 119 (see Table 10). The difference between the two anxiety score means for both sex and 08? were highly insiglificant, while the differences re- sulting from the experimental manipulations are significant at beyond the .001 level. For the to Control Ss there was an average decrease of .275 from Anx-I to Anx-II while the ho Mild—Aux Ss went up an average of .773 and the ho Hi-Anx Ss went up an average of 1.750. Thus the maniuplations spread the three groups almost equidistantly into one point intervals along the six point anxiety scale. The mean Amt-II score for the Control, Mild-Arm, and Hi-Anx groups were 1.725, 2.779, and 3.750, respectively. The second independent test of the success of our manipulations Table 10 Analysis—of-Variance of the Difference Betwaen Anx-I and Amt-II Scores Across the Treatment (A), Sex (B), and Ordinal Birth Position (C) Variables. Variance Source Sunrof Smrares _d_f_ Mean Sggare _F_ .Sigificance A 82.05000 2 b1.02500 5 3.901176 p (.001 B .03333 1 .03333 .0h380 - N.S. c . 30000 1 .30000 .39h16 N.s. AB .8166? 2 .1108 33 .53650 N.S . AC .05000 2 .02500 .03285 N.S. BC .0 3333 1 .03333 .014380 N.S. ABC 1.0166? 2 . 50833 .66788 N.S. Error 82.20000 108 .76111 Total 116.50000 119 involved the Ss' responses to the terminal anxiety question. When given the chance, 12 out of to Se in the Hi-Anx condition chose to terminate as compared to only 1 out of to in the mid-Aux condition (see Table 11). The difference between these frequencies is significant at the .02 level by Chi Square (of. Ferguson, 1959, p. 172). Given that the experimental manipulations worked as planned, two other methodological questions had to be considered prior to the testing of the mpotheses. The first involved the reliability of the instru- ments as they were employed in the present study. Consequently, the Ss' F and Dognatism scores on the two forms were correlated and the results for the major sub-groups are presented in Table 12. In the main, these correlation coefficients are slightly greater in magnitude Table 11 Chi Square Test of Responses to the Terminal Anxiety Question. Decision to Terminate in is. Mild-Am: 1 39 A0 Condition Hi Am: 12 28 b0 13 67 80 Chi Square = SJ-Ll, p = .02 Table 12 Reliability Coefficients for'Dognatism and F subscales: Dissertation Study 91 Dofltism _I‘: ran 3: corrected 5 raw 5 corrected 5 Total sample 120 .72582h .8141133 .699800 .823391 Males 60 .692817 .818538 .690972 .8172h8 Females 60 .756869 .861611 .710360 .830656 First and onlies 50 .736h92 .8h825 3 .675928 .8066 31 Later horns 60 .72h312 .8hon7 .716259 .831675 Control So to .7813h7 .87725‘1. .711961 .8 317h9 Mild-Am: Ss 1.0 .751975 .858h31 .7667 39 .867971 Hi—Anx 33 1.0 .5919h2 .7h3673 .shh015 .7014676 than those obtained in the earlier reliability study.. In fact, for the 56 to Control 83, the corrected reliabilities for the Dognatism and F- Scales were calculated to be in the order of .88 and .83, respectively. For the entire sample (i.e., experimental Ss included) the respective coefficients were .81; and .82. These values are even better than those obtained earlier in the protesting of the instruments (see Table 1.). Attention should also be directed to several additional points. First, the Doglatisn sub-scales appear to be slightly more reliable than the F sub-scales. Second, scores for the fenale Ss seen to be generally more consistent and reliable than those for the male Ss. Third, there appears to be a general trend for the reliability toqbe inversely re- lated to anxiety -- as situational anxiety increases, reliability de- creases. This last point is a rather interesting finding and‘will be discussed in greater detail below. The last methodological consideration concerns the possibility of constant form or order effects involved in the adninistration of the sub-scales to the 8s. To determine whether either, neither, or both were present, in addition to the data for the to Control So, the data. for the 17h Be used in the earlier pretest reliability study were also employed. (Inasmch as differences of unknown nayitude resulting from the erperinental nanipulations were predicted between first and second fern scores for the 80 oxperinental Ss, the data these So could not be used in testing for former order effects.) The data for the l7h reliability studySs (see Table 13)3 soar to indicate the presence of both constant form and order effects for 3. Note: these are some of the sane data previou81y presented in Tab .h th b ran inord tofacilitatee compiling Sitfi‘Thre‘Ihf“ rear god or soy Table 13 Protest Reliability Study: Dognatism and F subscale means. Total 3 1e Form A first Form ~13 first N II 17% N = 86 N = 88 Mean Dogaatisn Score: 72.603 73.291 71.932 Form A Mean Dognatisn Score: 75.371; 71)..th 76.2814 Form B Moan FfSCOI‘B: 1169.391 147.105 145.693 Form A Mean F-Score: 1.7.190 h7.2lu‘4 H.136 Form B ------------------------...-.M’----- both the F and Dogmatism sub-scales. With respect to a form effect we find that, when we consider the total sampleJ the average Form B Dogma- tism score is 2.8 points higher than the average Form A Dognatism score. Similarly, the average B Form F-score for the total sample is .8 of a point higher than the average A Form F-score. Earlier, in our dis- cussion of the construction of the sub-scales in Chapter III, we noted that while the difference between F sub-scale means was not statis- tically significant, the difference between Dognatism sub-scale means was significant at better than the .01 level. The pretest reliability study data also indicate an order effect. For both the F and Dogmatism sub-scales, the first form taken averages approximately one point higher than the total sample mean, while the second form taken averages a point less. When we consider the data on the 140 Control 55 in the present study there again seems to be a fairly constant form effect (see Table 1h). 58 Table 1h Dissertation Study (Control group): Dogeatisn and F subscale means. Total 3 le Form A first Form B first n 8 E0 n B 20 n a 20 Mean Dog-Latin Score: 69.600 72.600 66.600 Form A Mean Dognatisn Score: 71.550 75.350 67.750 Forn B Mean F-‘S'core: 140.700 18.600 37.800 Fora A Mean FLScore: h2.375 hs.700 39.050 Form B . For both F and Dopatisn sub-scales, the moan total ample B Fern score is approximately two points higher than the nean total sample A Form score. The order effects, however, appear to be more random. For the Dogaatisn sub-scales, if' Fem A is taken first the resultant score tends to be approximately three points higher than the total sample average. If Four B is taken first, it tends to average approximately four points less than that(i.e., total sample) average. For the F sub- scales,‘if Form A is taken first it tends to average approximately three points higher than the total sample mean. When Fem B is taken first it tends to average approximately three points less than that (i.e., total sanple) mean. 0n the basis of a rough average of the two studies, it appears that the. order effects are random. Consequently, no adjustment on these grounds was believed necessary. However, the data do seen to 59 indicate a constant form effect for both F and Dognatism sub-scales, viz., scores for Form B tend to average about two points more than scores on Form A. Consequently, a simple adjustment was made prior to testing the hypotheses. The correction entailed subtracting two points from the Form B (Dognatism and F) subscale scores just prior to the calculation of the difference scores. Thus, the difference scores (dX) used in the analyses of variance described directly below were com.- puted as follows: For all 55 who took Form A first: dX I (Form B score minus 2) minus Form A score For all Ss who took Form B first: 8.1! = Form A score minus (Form B score minus 2) These difference scores are presented in Table 15. Tables 16 and 17 contain the means and S.D.s of these difference scores for dognatism and authoritarianism, respectively. Two separate analyses-of—variance were then conducted. One tested the hypotheses relating to dognatism (Hypotheses Ia and It), and the other tested the hypotheses relating to authoritarianism (Hypotheses 11a and ID). The results of these analyses are presented in Tables 18 and 19. With respect to dognatisn, inspection of Table 18 reveals that no significant relationships are obtained as a result of either the experimental manipulations (i.e., anxiety treatments) or the organismic variables involved (i.e., sex and birth position). The results for authoritarianism (see Table 19) are similarly unencouraging. The only sigrificant difference found was for the sex variable. Males showed a simificantly greater increase in authoritarianism than did females 60 Table 15 n n. warriors? ntsiwatmu etmésinwn hormonzmnia . ) n m n.7nmr9mncvn nunqoegnn $33983, sharpener... m b m 0 m m 8 O .m 5 m3 lgnflouuiwndm. O27ababonw2|468 348n2flm‘mewfl7 7%M90ld86 H t .n m 3 m... ntraenhonio snosonnorm vivannmeen nflhnmohni H he a a no a an s s m m1 s saturation 22437162.... mnzvnchnno mgnnonsmin a 1 c - - . e... C m 0 r t n . 3 we a n. ...,oosnssin... sternum. .03 97mins??? gfimnmtnui m )8 C an m (a n ms at g u .. m . m m h m. m o r m & B & B .n t. u ..o r .W .m .m .3 m 8 u ..u a F h F Q h l a n m m_ m m C F C 8e 61 Table 16 Means and S.D.s of the Difference Scores for Dogmatisn.a Treatment fig; Control Mild-Ana: HieAnx ”13 Control Control MiIdAAnx Mild-An: Hi—Anx Hi-Anx Control Control MildAA‘nx Mid-An: BieAnx Hi-Anx lllil-‘Il ’113’133’113 C0ntrol C0ntr01 control Control E5 mezzeezzees: eon: a. A blank in a column indicates a breakdown with no concern for that particular variable. Birth renhb Early berns Later barns Early borns Later borne Early borne Later borne Early berns Later borns Early borns Later horns Early borne Later borne Early horns Later borns Early borne Later horns Early borne Later horns Early borne Later borne Early borne Later borne Early borne Later borns 7! ho ho to 60 6O 60 60 20 20 20 20 20 20 20 20 20 20 20 20 fi fi 30 30 10 10 10 10 10 10 10 10 10 10 10 10 120 Mean .800 1.650 2.700 1.583 1.850 2.750 .683 4.500 3.100 2.250 1.050 h.000 1.100 3.150 -1.550 1.350 1.950 3.750 1.650 1.900 1.267 3.600 .100 .hoo “Behoo 5.900 .300 1.900 2.600 .800 1. 3.338 13.600 h.100 4.300 1.717 b. "Early borne“ signifies first and only home. S.D. 9.14 3 10. 5 10.118 10. 317 9.377 8.08h 10.711 8.161.: 9.99h 11.182 9.017 10.829 9J50 7.680 10.585 7.271 12.h12 11.206 9.069 8.1m 12.066 9.2h2 9.331; 5.777 10.h80 8.621 10.91h 6.657 lh.81l 8.162 10.2h2 12.057 10.069 10.929 7.273 9.818 62 Table 17 Means and S.D.s of the Difference Scores for Authoritarianism.a Treatment pg Birth rankb _Ig Mean 5 .D. Control - .... to .225 9.690 Mild-Am: - .— ho 2.h00 9.108 Hip-Am: - - to 3.925 9.937 .. M .. 60 b.600 8.857 .. F .... 60 -.100 9.832 .... -'- Early borne 60 1.350 10.183 - — Later horns 60 3.150 9.001 control M ... 20 3.350 5.373 Control F — 20 -2.500 12.077 mm M .. 20 h. 750 9.1108 Mild-Ana: F .... 20 .050 8.376 Hie-Aux n .. 20 5.700 11.103 Control - Early bores 20 -.100 9'. 358 Control - Later horns 20 .950 10.266 , Hilda-Ann - Early horns 20 1.100 11.253 Mild-Jinx - Later borne 20 3. 700 6. 33h Hip-Aux - Early borne 20 .050 10.123 W - Liar b0“ 20 e800 9e929 .. M Early horns 30 3.h33 9.877 .... M Later borns 30 5.767 7.695 — F Early home 30 -.733 10.218 .... F Later borns 30 .533 9.562 control M Early horns 10 3.h00 6. 381 control 11 Later borns 10 3.300 h.h98 Control F Early borne 10 -'-3.600 10.811: Control F Later horns 10 4.1100 13.721 Mild-Jinx 11 Early borne 10 h.h00 11.796 Mild-.Anx 1’! Later borns 10 5.100 6.887 Mild—An: F Early barns 10 -2.200 10.196 Mild-Am: F Later borne 10 2.300 5.736 Hi-Anx M Early horns 10 2.500 11.511 W M Later home 10 8.900 10.216 Hi—Anx F Early berns 10 3.600 9.119 Hie-Ame F Later borns 10 .700 8.111 _ - .. 120 2.250 9.612 a. A blank in a column indicates a breakdown with no concern for that particular variable. b. "Early borne" signifies first and only horns. 63 Table 18 Analysis-of-Variance for the Difference in Dognatism Scores ‘ on the two Forms as a Function of Anxiety Level (A), Sex (B), and Ordinal Birth Position (C). Source Sum of Squares if Mean Square 3: Sigificance A 72eh67 2 $0233 efiS NeSe B 2.133 1 2.133 .022 N.S. C 128.133 1 128.133 1.h68 N.S. AB 291.h67 2 1&5.733 1.291 N.S. AC 1h0.h67 2 70.233 .708 N.S. BC 61.633 1 61.633 .621 N.S. ABC. 55.167 2 27.733 .279 N.s. Error 10718.600 108 99.2h6 Total 11h70.367 119 Table 19 AnaIySi‘se-of-Variance for the Difference in FAScores on the two Forms as a Function of Anxiety Level (A), Sex (B), and Ordinal Birth Position (C). m Sum .of‘A‘Squares 9.}; Mean Square . E Sigrififlpc A 2A6.350 2 123.175 1.37h N.S. B 662.700 1 662.700 7.391 p<.0l C 97.200 1 97.200 .lh7 N.S. AB 26.h50 2 13.225 1.08h N.S. AC 12.050 2 6.025 .067 N.S. BC_ 8.533 1 8.533 .095 N.S. ABC 257.017 2 128.508 l.h33 N.S. Error 968h.200 108 89.669 Total 10991:. 500 119 61: (for 1 and 108 df; F :3 7J4, p<.01; see Suedecor, 1956). This result is difficult to interpret for two reasons: (a) there were no a priori hypotheses regarding a sex variable effect, and (b) any sex variable effect obtained was expected to be meaningful only in toms of its interaction with the other variables (i.e., anxiety and ordinal birth position). Close examination of Tables 16 and 17, however, reveals some rather interesting (though non-significant) things. Confining our in- terest first only to Table 16 we find that, as predicted, there is an increase in cremation with increasing anxiety. For the to Control 33 there was an unpredicted mean increase of .8 of a point. However, in line with our predictions, we find a mean increase of 1.650 r... the ho PfildnAnx Se, and a mean increase of 2.700 for the ho Hi—Anx Ss. We also find, in line with our predictions, that while later home only display a mean increase of .683 of a point, first and only borne ex- hibit a mean increase of 2.750 points (p = .12). Furthermore, the interactions also generally conform to prediction. Later born Control Ss who were predicted to show the least amount of increase actually displayed a mean less of 1.550 points, while anxious first and only born 33 displayed the greatest mean increase, 3.750 points, as pre- dicted (p = .06). The relationships between birth position, annety level, and change in dogmtisn are presented graphically in Figure 2. Although these differences are all in the directions predicted, they apparently fail to reach statistically significant levels for two reasons: (1) If one considers that the range of possible scores obtainable on each 20 item subscale is 20 to 11.0, then a change of two _——.—~__._z_ _ _ 65 +h-T-——~~ ~ I , ..- -. ..MW 7 ( i £irst and 0 ; ; ohly borns ! .- 4.» I , ; +3¢ - — . -.-.l.- . r . - . , . ..-. .- ____§ ' 2 ' 3 ? ' i I I .. i ? § 8 .2... , . , .--—4 E r; r ' v Later borns' U u i : +1?“- L |I .. 1 Change ; i .12. s f o T ...”---lflu -- - ... ,..7 - .- .. -- in- .. _ -l .... Score I g . i ' s l l i c . I 9 .1 A _1: ...- .-...- :---. .. -.. ...- ' t i . ' ’ . ‘ ' . A ' ; ’ , .I . g g . p ‘ ' I ‘ I l L l ; -. -' .-. ---—.—_———, I I o l l he--- 1.- . Control Se Milqunx Ss HieAnx Ss I— Fig. 2. Mean Change in Dognstisn.scores obtained for first and only borne vs. later borne in each of the three treatment groups. Note: With each questionnaire item based on a l-to-7 scale. The range of possible scores attainable on either of the 20 item Dognatism sub- scales is 20 to 1&0. Therefore, the changes in score possible when going frat one form to the other can range. anywhere from 0 to a maximum of (lho - 20 I) 120. Even fer a hypothetical individual who responded neutrally to each item on the first form (thereby Obtaining a score of 20 X h I 80) , the range of change possible in either a positive or negative direction would still be from 0 to 60. Thus, the small changes presented in the above figure must be interpreted in this context. 66 or so points seems rather small. ( 2) Examination of the raw differ- ence scores (Table 15). the standard deviations (Table 16), and the error mean square of the analysis of variance (Table 18) readily re- veals a great deal of variability among the scores. Table 17 also reveals small increases in the predicted direction for authoritarianism resulting from increases in situational threat. While the Control Ss demonstrated an unpredicted increase of .h25, the Mild-Aux and £11qu Ss demonstrated respective predicted (but non- sigrificant) increases of 2.1:00 and 3.925 points. However, contrary to prediction, later born Ss evidenced a greater increase (3.150 points) than did the first and only born Ss (1.350 points). The relationships between birth position, annety level, and change in authoritarianism are presented graphically in Figure 3. Again, we note the following: (1) When the range of possible scores obtainable on each 1h iten sub- scale is lh to 98, a change of two or so points appears relatively small. (2) icanination of the raw scores (Table 15), the standard deviations (Table 17), and the error mean square of the analysis of . variance (Table 1?) reveals a great anount 0f “01PM“ 7013’111'07 among the difference scores. A possible factor contributing to these findings may arise from the relatively lower reliabilities of the Domtism and F subscales for the Iii-Ann condition (cf., Table 12). Since the statistics employed assume that the reliability coefficients are all estimates of the same population parameter, Edwards (1960, p. 83) test of homogeneity for L: values of _r; was applied to the raw data. For the three Dognatisn sub- scale treatment group correlation coefficients (i.e., .78131t7, .751975, r.v‘ I l II III! III I I'll .... 67 U I ”...... w‘ ‘- - t i ‘ 7 | .' L --..-W._--.4... . . . ; 7‘ -- , 1. . ...---..._- W W W Pig. 3. Mean Change in Authoritarianisn scores obtained for first and only borne vs. later borne in each of the three treatment groups. l Note: With each questionnaire item based on a l-to-T scale, the range of possible scores obtainable on either of the 11: item "1'" sub-scales is 1h to 98. Therefore, the changes in score possible when going fraa one fan to the other can range anywhere fran 0 to a maximal of (98 - 1h 3) 8h. Even for a hypothetical individual who responded neutrally to each item on the first form (thereby obtaining a score of k X 11: II 56) , the range of change possible in either a positive or negative direction- would be frm 0 to 1&2. Thus, the small changes presented in the above figure must be interpreted in this context. 68 and .59l9h2) the obtained Chi Square was 2.858. With 1o-1-2 df, this fails by a substantial margin to approach significance (p = .25, ca.). Analyses of the three F subscale treatment group correlation coeffi- cients (i.e., .711961, .766739, and 51414015) yields 8. Chi Square of 3.1m. with k-l = 2 degrees of freedom, this also fails by a sub- stantial margin ( p = .22, ca.) to reach significance. Lastly, it should be noted that anxiety may be defined from either of two perspectives -- that of the experimenter or that of the subject. In the present study, as well as in the analyses presented thus far, we have employed the vantage point of the experimenter. Depending upon which of the treatments he was subjected to, a subject was considered to be either mildly anxious, highly anxious, or not aanous at all. One could say that anxiety was here being operationally defined by experimenter fiat. However, it is generally recognized that any two individuals selected at random and exposed to exactly the same manipu- lations might differ widely in their reactions. Thus, in some experi- ments involving the measurement of anxiety it is the subject who defines his emotional condition. This is usually done either through some sort of verbal report or by measurement of the subject's physiological re- actions. In this study, "verbal report" data aroavailable for each subject, viz. , the Aux-I and Aux-II scores, and the responses to the terminal anxiety question. Consequently, it was decided to re-axamine the change in Doguatism and Foe cores using these self-appraisals as our measure of anxiety. In one such analysis the 80 emerimental Ss were grouped on the basis of their Amt-II scores. These scores represent the S's self- 69 reported level of anxiety subsequent to the experimental manipulations. Simple visual inspection of this data array (see Table 20) revealed neither consistent nor meaningful trends. A similar arrangement, in which the Dogmatism and F change scores of the 80 experimental 83 were grouped according to the pre- to post-manipulation change in the 83' anxiety level (i.e., Aux-II score minus Aux-I score), also proved fruitless. In yet another analysis, the ho Ss who had undergone the Hi—Anx manipulations were arranged on the basis of their responses to the terminal anxiety question (see Tables 21 and 22). Certainly, one could legitimately classify as being truly "highly anxious" those 12 88 who, after undergoing the HinAnx manipulations, had responded in the affirmative when asked "Would you like to leave rather than proceed with the experiment?" is opposed to the 28 Se who responded negatively to this question, these 12 were ready to give behavioral expression to their verbal report of high anxiety. Although _t tests indicated no significant differences between any of the major or sub groups, it is interesting to note that the five "highly anxious" first-born termin- aters, as predicted, showed by far the greatest mean increase in both Depletion and F-ecores. The failure to obtain siglificance again appears to be due to the higher variability within the scores. 70 Table 20 Mean Changes in Domatism and F Scores. (Where the 80 experimental subjects are grouped according to their Anx-II Scores .) hm) ANX Ss (xi-31:0) ANX-II SCORE DOGMATISH E 1_\I S First Borns 0.00 0.00 0 Later Borns 10.00 14.00 l h First Borns -1.75 1.25 h Later Borns 18.00 0.00 l 3 First Borns -0.70 7.3) 10 Later Borns 1.60 5.60 10 2 First Boms 6.83 -9.33 6 Later Barns 0.67 1.33 6 1 First Borns 0.00 0.00 0 Later Borns -h.50 3.00 2 HI ANX Ss (who; ANX-II scan Doommsu F. _13 6 First Borne 11.50 12.00 2 Later Bone h.67 1h.oo 3 5 First Boms -1.17 h.33 6 Later Boms -1.50 -S.33 3 h First Borne 5.50 5.00 2 “tor B01313 0.00 -2.00 3 3 First Borne 1.h2 0.85 7 Later Borns h.20 6.h0 10 2 First Borns 8.33 -l.67 3 Later Boms 0.00 0.00 0 1 First Boms 0.00 0.00 0 Later Borns -9.00 20.00 1 71 Table 21 Mean Increases in Domatism Scores for Hi-Anx Treatment Subjects. (Where the "Terminator" vs. "Non-Terminator" distinction refers to the individual's response to the terminal anxiety question.) Tammroas NON—TERMIxmoag . Y = 9.00 Y = 2.93 i = h.h5 F1432 Boms SeDe 3 15095 S.De : 10056 SeDe = 11e65 N = 5 N =15 N :20 "i = 0.11; '2' = 2.116 T: = 1.65 Later Borne S.D. = 8.75 S.D. = 9.148 S.D. :- 9.00 _"—' N = 7 N =13 N =20 SE = 3.83 if = 2.71 S.De =11es9 SeDe = 9e89 N =12 N :28 Table 22 Mean Increases in F-Scores for Hi-Anx Treatment Subjects. (Where the "Terminator" vs "Non-Terminator" distinction refers to the individual's response to the terninal anxiety question.) . TWTORS NON-TWOIE '2, = 13.00 2' Z 0.07 i = 3.30 Fiat Borg S.D.: 12e79 S.D.: 6efi SeDe = 8.02 N = S N :- 15 N = 20 i = 0.57 T! = 6.69 'i = 5.50 Ear B01113 SeDe = 10e67 30130 = 9013 Sons 3 9eh2 N 3 7 N 3 13 N 3 20 .2 = 5e75 i = 30114 S.D.: 11002 SeDe = 7.62 N = 12 N = 28 Chapter V: Discussion and Conclusions Although there were some small trends in the predicted directions, no statistically significant support was obtained for any of the hypo- theses. Several explanations may be offered to account for this failure. One distinct possibility is that the hypothesized relation- ships have no basis in reality, i.e., they just do not exist. Another alternative is that they are correct as stated but were, for one reason or another, improperly tested, i.e., the experimental manipulations and/or measurement procedures were in error. A third possibility is that the hypotheses are, in general, conceptually accurate but are lacking or in error in some of their essential details. There is some evidence (viz. , Rokeach, Toch, and Rottman, 1960) to indicate that, for Hypothesis Ia at least (i.e., contraction of the belief system will vary directly with the degree of anxiety elicited by the situation), the first possibility can be discounted. The second possiblity would also appear to be ruled out for at least two reasons: (a) almost identical threat manipulations and measurement techniques have been successfully employed in several other studies (e.g., Schachter, 1959; Gerard and Rabbis, 1961; Sarnoff and Zimbardo, 1961; and, Staples and Walters, 1961); and, (b) two independent tests of the efficacy of the present anxiety inducing procedures resulted in con- firmation at highly statistically sigiificant levels. 0f the possi- bilities considered, the third, therefore, seems to offer the most likely emlanation. The primary independent variables of this study are ordinal birth 72 73 position and situational threat or anxiety. or the two, ordinal birth position is the more difficult to confound, i.e., one either is or is not an only born, a first born, or a later born. In only a small minority of cases, such as when 53 are adopted children or when parents re-marry and provide the S with half-siblings, is there a problem of categorizing into first and later home. The threat-anxiety variable, however, is a bit more difficult to work with, primarily because psych010gists have not yet agreed upon any single universally acceptable definition of anxiety. For one group of psychologists, anxiety and fear are two distinct affective states. Generally speaking, the term £333 is applied to those cases where the object of danger is clearly perceived, and the term anxiety used when the object is unknown or vaguely discerned (cf. , Symonds, 19146). Many of the theorists in this camp assert that the cardinal difference between the two is to be found at the most funda- mental levels of the personality structure. A very frequently cited definition of anxiety (May, 1950) holds that it is a diffuse "apprehen- sion cued off by a threat to some value which the individual holds essential to his existence as a personality." Portnoy (1959, p. 309) has made the following distinction: Anxiety is a natural phenomenon which the individual experiences when values essential to his existence, his sense of being, and his identity are threatened. It is to be distinguished from fear in which the threat is peripheral, the intactness of the sense of being is not being threatened, the danger is objective, and the in- dividual can evaluate it and can act either in terms of fight or flight in coping with it. Other investigators, though maintaining the distinction between fear and anxiety, conceptualize the difference at less fundamental 7h levels. Cameron (19h?) defines fear as "any strongly avoidant emotion- al reaction that culminates in flight when flight is possible" (p. 1146) and anxiety as "the normal preliminary phase of emotional flight, but which for some reason is prevented from going into its consummatory phase" (p. 1&7). For Cameron, then, anxiety is similar to the early stages of fear. However, in another conception (Goldstein, 1939), fear is conceived of as the early stage of anxiety. What is it then the leads to fear? Nothing but the emerience of the possibility of the onset of musty. What we fear is the impending anxiety. Thus it be- comes clear that anxiety cannot be made intelligible from the phenomenon of fear, but that only the opposite procedure is logical. The person in fear knows anxiety from past experience and present imagination (anticipation). The person in anxiety, however, cannot ~ know fear, because in the state of anxiety he is in- capable of any recollection. The person in fear infers, from certain indications, that an object is apt to bring him into a situation of anxiety. (Goldstoin. 1939. pp- 296-297). Briefly we note that other definitions popular at one time or another have held anxiety to be: (1) an oven-generalisation of the fear response (Mowrer, 1939) g and ( 2) the consequence of inadequate control of fear (Grinker, 1939)- . lastly, there are those who assert that "nothing is gained ... by making a systematic distinction between anxiety and £295 .... Such distinctions are more linguistic than psychological. Whatever the status of the amusing object, the basic emotional reaction is the same. In the literature on neurosis, anxiety is the term most often encount- ered, but it is used in a sense that includes all degrees of the fear reaction" (White, 1956, p. 206). Hall and Lindsey (1951;, p. 160) state simply that, insofar as psychoanalytic theory is concerned, "Anxiety 75 and fear are interchangeable terms." Physiological psychologists would concur with this position and would point to the general failure thus far to demonstrate any heme-physiological differences between the two. It thus appears that the investigator, confronted by an array of interpretations, can literally select a definition of anxiety suitable for his purposes. In the formulation of the present study it was be- lieved that no distinction needed to be drawn between fear and anxiety, Two considerations led to this decision. For one, several other studies in which birth position was the variable of interest (e.g., Schachter, 1959: Gerard and Rabbie, 1961’; Sarnoff and Zimbardo, 1961; Staples and Walters, 196l; and, Weller, 1962) employed very similar manipulations (i.e., threat of electrical shock) and defined the affective consequences as anxiety. For another, the statement from which the primary hypothesis of this study is derived (viz., ”we assume that the more threatening a situation is to a person, the more closed his belief system will tend to become," Rokeach, Toch, and Rattan, 1960, p. 377) specifies threat, not necessarily fear nor anxiety. Consequently, it was assumed that fear manipulations, even if dis- criminably different from anxiety manipulations, would serve equally as well to produce a "threatening situation." However, if we accept the definition of anxiety proposed by May (l950) and Portnoy (1959), and if we re-examine the earlier test of the Impothesis concerning belief system contraction as a function of situa- tional threat (i.e., Rokeaoh, Toch, and Rottman, 1960), we find that a case can be made for an anxiety (vs. a fear) interpretation of the phrase "threatening situation". It will be recalled that in their test 76 of this proposition these authors consider the calling of an ecumenical council to be indicative of a threatening situation. They document their belief by noting that: According to the Catholic Engclopedia (1908), ecumenical councils represent "a common effort of the church, or part of the church, for self-preservation and self-defense. They appear...whenever faith or morals or disci line are seriously threatened" (p. 1421;). [italics ours If one accepts the fear-anxiety distinction proposed by May (1950) and Portnoy (1959), it then becomes tenable to assert that what was invol- ved in the Rokeach, Toch, and Rottman study was the societal-level counterpart of individual anxiety, as opposed to individual fear. Let us pursue this point further. One of the most generally accepted and frequently used measures of anxiety (as a personality trait) is the Taylor Hanifest Anxiety Scale (TMAS). If one assumes that the TEAS is a valid indicator of anxiety, insofar as shock or threat of shock is anxiety inducing, one would expect individuals scoring high on the M to react differently than low-scoring indivi- duals, i.e., there should be a relationship between personality measures of anxiety and physiological indices of autonomic nervous systea areusal. However, it appears as if such studies have consistently obtained negative findings (cf., Hodges and Spielberger, 1%6 , p. 288). For example, stisohn (1956) found changes in heart rate in response to actual shock to be uncorrelated with scores on the THAS. Similarly, Katkin (1965) found that in response to threat of shock, changes in skin conductance and the BBB were unrelated to MS scores. the way to interpret these findings is suggested by Spielberger's recent (in press) conceptualization of anxiety, in which a distinction III .lidql lllll Ill-l. .-J 11 l Ills-II I0l1: I ll ' I I l 77 is made between state- and trait-anodety. "According to this view, state-anxiety is a transitory condition of the organism characterized by subjective feelings of apprehension and heightened ANS arousal... whereas trait—anxiety is a relatively permanent aspect of personality which reflects individual differences in the disposition to respond with state-anxiety to situations perceived as threatening" (Hodges and Spielberger, 1966, p. 293). Rophrasing the issue in the present study, one could say the following. Although dognatisn has already been shown to be positively related to, and a partial function of, trait- anxiety, the current investigation sought to demonstrate that domatisn could also be affected by state-mxiety. According to Spielberger, 'Ss who differ in trait-anxiety will respond with differential aaounte of state-anxiety to 'ogo-stress' situations, but not to situations ine- volving physical pain or threatof pain" (Hodges and Spielberger, 1966, p. 293). This contention is entirely consonant with the results of the present study. If the reasoning presented here is correct, then two ramifications emanate from such a position. The more imediate one is that the hypo- theses were not adequately tested by the present study, i.e., fear, rather than the discrininably distinct state of anxiety, was unployed as the independent variable. The other isthat Rokeachian theory is in need of a slight but very important modification, namely, the specifi- cation of anxiety as the type of threat necessary to cause the contrac- tion of a belief system. Thus, it is now obvious that certain procedural modifications must be incorporated in any future tests of the Impotheses. Specifically, 78 the threat manipulations should attempt to induce affective reactions which more nearly approximate anxiety rather than fear, as the two terms are distinguished in the formulations of May (1950) and Portnoy (1959) . We believe that such manipulations would be in greater conson- ance with Rokeach's implicit conception of threat (i.e., threat qua anxiety), and would provide a more adequate test of that facet of his theory. To produce such anxiety it is necessary that the threat be pemeived by the subject as being vague and ill-defined. Further, it must be directed towards the deepest personality levels -- those in- volving the individual‘s core concepts regarding self, being, identity, and existence. (In contradistinction, in the present study the threat was concrete, objective, delimited solely to the infliction of external physical pain, and peripheral to the individual's self-concept.) To actually obtain state-anxiety via experimental manipulations would appear to be a fairly difficult task. For example, although there are general and specifiable fear stimuli (e.g., probably all normal Americans are afraid of severe pain), with the possible excep- tion of death, the writer is not aware of any general and specifiable anxiety stimli. Furthermore, since the referent for fear is ob- Jective, the experimenter can either directly control it or, even at the most basic level of scientific methodology, he can observe its onset, duration, and cessation and the individual's attendant re- actions. However, since anxiety is not directly contingent upon the external stimlus, even if the investigator could determine that the cause of the anxiety had ceased, could he also safely assume that the annety itself had dissipated? This ties in with the conception of 79 anxiety as being directly involved with the deepest levels of the in- dividual‘s personality, as opposed to fear which is involved with the intermediate and/or more superficial levels. Lastly, the writer wishes to pose two related questions which he does not at this time feel qualified to answer himself: (1) If, in order to induce state- anxiety, we mst manipulate the individual's deepest levels of person- ality, should such research be conducted only with the active presence and assistance of a clinician or psychiatrist? (2) Ethically speaking, should such research even be conducted at all? On the basis of the current investigation the following statements appear in order: 1. Early and later borns enter the experimental situation ex- periencing approximately the same degree of general nervous tension. (Note: there are some grounds for considering the Anx-I score, since it was administered prior to any specific threat manipulations, to represent a valid measure of pro-experimental anxiety.) 2. Neither dognatism nor authoritarianism is directly related to ordinal birth position, either under fear or non-fear inducing condi- tions. 3. Situational threat, here being operationally defined in terms of fear manipulations, has no significant effect upon either dognatism or authoritarianism. 1:. Rokeach's (1960) theory appears to be in need of a minor re- finement, i.e., the nature of the threat theorized to affect dogmatism should be more specifically defined as anxiety rather than fear. 80 Insofar as the current study failed to employ true anxiety manipulations, no conclusions may be offered regarding any of the hypothesized relationships. 81 References Adorno, T. W., Frenkel-Brunswik, Else, Levinson, D. J., & Sanford, R.N. The Authoritarian Personality. New York: Harper & Row, 1950. Ax, A. F. The physiological differentiation between fear and anger in humans. Psychoscms Med., 1953, 15, h33-hh2. Becker. S.U., and Carrol, Jean. Ordinal position and conformity. J. abnom. soc. szchol" 1962, 65, 129-131. Becker, S. W., Lerner, M. J., & Carrol, Jean. Conformity as a function of birth order, payoff, and type of group pressure. J. abnormi soc. ngchol., 196h, 69, 318-323. Becker, S.'H., Lerner, M. J., and Carrol, Jean. Conformity as a function of birth order and type of group pressure. J. pgrs. soc. Psychol., 1966, 3, 2h2-2hh. Cameron, N. The pgzchclcgy of behavior disorders. Cambridge: The Riverside Press, 19 7. Campbell, A, Factors associated with attitudes toward Jews. In G. E. Swanson, T. H. Newccmb, a E. L. Hartley (Eds.), Readings in Social Pszghglggy_(Rev. ed.). New York: Holt, 1952. Capra, P. C., and Dittes, J. E. Birth order as a selective fact r among volunteer subjects. J. abnorms soc. P hol., 1962, 6 , 302. Catholic Enczglopgdig. New York: Gilmsry Society, V01. IV, 1908. Centers, R. The pszghology of social classes. Princeton: Princeton University Press, 19 9. Centers, R. The American class structure:‘ a psychological analysis. In G. E. Swanson, T. M; lewecmb, a E. L. Hartley (Eds.), Readings in social pgzghology (Rev. ed.). new York: Holt, 1952. Christie, R., and Cook, Peggy. A.guide to published literature relating to the authoritarian personality through 1956. J. Psychol., 1958, 16 . 171-199 . Christie, R., Havel, Joan, & Seidenberg, B. Is the F Scale irreversi— ble? J. abnorm. soc. Psychol., 1958, 56, 1h3-159. Davids, A. Some personality and intellectual correlates of intolerance of ambiguity. J. abnormt soc. Psyphol., 1955, 51, his-520. Davids, A. The influence of ego-involvement on relations between authoritarianism and intolerance of ambiguity. J. consult. Psychol. 1956, 20, 179-18h. 82 Davis, A. Socialization and adolescent personality. In G. E. Swanson, T. M. Newcomb, & E. L. Hartley (Eds.), Readings in social ggychol- 25; (Rev. ed.). New York: Holt, 1952. Deane, G. E. Human heart rate responses during experimentally induced anxiety. J. e52. Psychol., 1961, 61, h89-h93. Deutsch, M. and Gerard H. B. A study of normative and infOrmational social influence upon individual Judgement. J. abnorm. soc. Psychol., 1955. 51, 629-636. Dittes, J. E. Birth order and vulnerability to differences in acceptance. Amer. Psychol., 1962, 17, 329. Dittes, J. E. and Capra, P. C. Affiliation: comparability or com- patibility. Amer. Psychol., 1962, 17, 329. Dollard, J., Doob, L. W., Miller, N. E., Mowrer, 0. H., & Sears, R. R. Frustration and aggression. New Haven: Yale university Press, 1939. Edwards, A. L. Experimental design ingpsycholOgical research. (Rev. ed.) New York: Holt, Rinehart, & Winston, 1960. Elizur, A. Content analysis of the Rorschach with regard to anxiety and hostility. J. prod. Tech., 19h9, 3, 2&7-28h. Ferguson, G. A. Statistical analysis in psychology and education. New York: McGraw-Hill, 1959. Fillenbaum, S. and Jackman, A. Dogmatism and anxiety in relation to prdblem solving: An extension of Rokeach's results. J. abnorm. soc. Psychol., 1961, 63, 212-21h. Frenkel-Brunswik, Else. A study of prejudice in children. Hum. Relat., 19h8, 19 295-306. Frenkel-Brunswik, Else. Social research and the problem of values: A reply. J. abnorm. soc. Psychol., l95h, A9, h66-h71. Fromm, E. Escape from freedqg, New York: Rinehart, 19hl. Fruchter, B., Rokeach M., & Novak, E. G. A factorial study of dOgma- tism, opinionation, and related scales. Psychol. Rep., h, 19-22. Gerard, R. B., and Rabbie, J. M. Fear and social comparison. J. abnorm. soc. Psychol., 1961, 62, 586-592. Goldstein, K. The organism, New York: American Book Co., 1939. 83 Greenberg, H., Guerino, Rosemarie, Lashen, Marilyn, Mayer D., & Piskowski, Dorothy. Order of birth as a determinant of personality and attitudinal characteristics. J. soc. Psychol., 1963, 60, 221- 230. Grinker, R. R. Hypothalmic functions in psychosomatic interrelations. mChOSOM. Made, 1939, 10 19-1970 Hall, C. S. and Lindzey, G. Psychoanalytic theory and its applications in the social sciences. In G. Lindsey (Ed.), Handbook of social psychology. Reading, Mass.: Addison-Wesley, 1953. Pp. 153-180. Hodges, N. F. & Spielberger, C. D., The effects of threat of shock on heart rate for subjects who differ in manifest anxiety and fear of shock. Psychophysiol., 1966, 2, 287-29h. Horney, Karen. The neurotic personality of our time. New York: Norton, 1937. Kardiner, A. The psychological frontiers of society. New York: McGraw-Hill, 1936. Katkin, E. S. The relationship between manifest anxiety and two indices of autonomic response to stress. J.,pers. soc. Psychol., 1965, 2, Kerlinger, F. Unpublished data. 196%. Lewinsohn, P. M. Some individual differences in physiological re- activity to stress. J. comp. physiol. Psychol., 1956, h9, 271-277. Lynd, R. S. and Lynd, H. M. Middletown. New York: Harcourt Brace, 1929. Lynd, R. S. and Lynd, H. M. Middletown in transition. New York: Harcourt Brace, 1937. Masling, M. New neurotic is the authoritarian? J. abnorm. soc. Psychol., l95h, h9, 316-318. May, R. The meaning of anxiety, New York: Ronald Press, 1950. Melikian, L. H. Some correlates of authoritarianism in two cultural groups. J. Psychol., 1956, h2, 237-2h8. Mbore, Rosemarie K. Susceptibility to hypnosis and susceptibility to social influence. J. abnorm. soc. Psychol., 196k, 68, 282-29h. Mowrer, 0. H. A stimulus-response analysis of anxiety and its role as a reinforcing agent, Psychol. Rev., 1939, h6, 553—565. 8h Pannes, Ernestine D. An exploratory study of the relationship between attitudes toward self and the degree of dogmatism present in a group of Junior-senior high school students. Unpublished doctoral dissertation, New York university, 1962. Powell, F. A. Open- and closedemindedness and the ability to differen- tiate source and message. J. abnorm. soc. Psychol., 1962, 65, 61-6h. Radloff, R. Opinion evaluation and affiliation. J. abnorm. soc. Psychol., 1961, 62, 578-585. Rokeach, M. The nature and meaning of dogmatism. Psychol. Rev., 195k, 61, 191‘ -201; e Rokeach, M. Political and religious dogmatism: An alternative to the authoritarian personality. Psychol. Monogr., 1956, 70, No. 18, (Whole No. h25). Rokeach, M. The open and closed mind: Investigations into the nature of belief systems and personality systems. New York: Basic Books, 1960. Rokeach, M. and Fruchter, B. A factorial study of dogmatism and re- lated concepts. J. abnorm. soc. Psychol., 1956, 53, 356-360. Rokeach, M. and Kemp, C. G. Open and closed systems in relation to anxiety and childhood experience. In M. Rokeach, The open and closed mind. New York: Basic Books, 1960, Pp. 3h7-365. Rokeach, M. & Mezei, L. Race and shared belief as factors in social choice. Science, 1966, 151, 167-172. Rokeach, M. and Rothman, G. The principle of belief congruence and the congruity principle as models of cognitive interaction. Psychol. Rev., 1965, 72, 128-1h2. Rokeach, M. Smith, Patricia W., & Evans, R. I. Two kinds of prejudice or one? In M. Rokeach, The open and closed mind. New York: Basic Book, 1960. Pp. 132-168. Rokeach, M., Toch, H. H., & Rottman, T. The effect of threat on the dogmatization of Catholicism. In M. Rokeach. The 0pgn and closed mind. New York: Basic Books, 1960. Pp. 376-388. Rosenfeld, H. Relationships of ordinal position to affiliation and achievement motives: Direction and generality. Unpub. report, 196k. Sampson, E. E. Birth order, need achievement, and conformity. J. abnorm. soc. Psychol., 1962, 6k, 155-159. Sampson, E. E. The study of ordinal position: antecedents and outcomes. In B. A. Mbher (Ed.), Progress in experimental;personality research, vol. 2, New York: Academic Press, 1965. 85 Sampson, E. E., and Hancock, F. T. Ordinal position, socialization, personality development, and conformity. Uhpub. NIMH Grant (14.5-71.7) , 1962. Sarnoff, I., and Zimbardo, P. Anxiety, fear, and social affiliation. J. abnorm. soc. Psychol., 1961, 62, 155—159. Schachter, J. Pain, fear, and anger in hypertensives and normotensives: a psychOphysiological study. ngchgsom. Mgd,, 1957. 19: 17-29. Schachter, S. The psychology of affiliation. Stanford, Calif.: Stanford university Press, 1959. Schachter, S. Birth order and sociometric choice. J. abnorm. soc. Psychol., 196k, 68, h53-h56. Schooler, C. Birth order and hospitalization for schizophrenia. J. abnorm. soc. Psychol., 196k, 69, 57h-579. Shils, E. A. Authoritarianism: "righ " and "left". In R. Christie, and M. Jahoda (Eds.), Studies in the scope and method of "The Authoritarian Personality" Glencoe, Illinois: Free Press,7195h. PP- 211-13- Siegel S. Certain determinants and correlates of authoritarianism. Genet. Psychol. Monoggg, 195b, h9, 187-229. Singer, R. D., and Feshbach, S. Some relationships between manifest anxiety,authoritarian tendencies, and modes of reaction to frustration. J. abnorm. soc. Psychol., 1959, 59, h0h-h08. Snedecor, G. W. Statistical methods, (5th ed.). Ames, Iowa: Iowa State College Press, 1956. Spielberger, C. D. Theory and research on anxiety. In C. D. Spielberger (Ed.), Anxiety and behavior. New York: Academic Press, in press. Staples, F. R., and Walters, R. H. Anxiety, birth order, and susceptibility to social influence. J. abnorm. soc. Psychol., 1961, 62, 716-719. Stone, L. J., and Church, J. Childhood and adolescence. New York: Random House, 1957. Stotland, E., and Dunn, R. E. Identification, "oppositeness," authoritarianism, self-esteem, and birth order. Psychol. Monog£., 1962, 76, 9, (Whole No. 528). Stotland, E., and walsh, J. A. Birth order and an experimental study of empathy. J. abnorm. soc. Psychol., 1963, 66, 610-61h. 86 Symonds, P. M. The dynamics of human adjustment. New York: Appleton- Century-Crofts , 1936. Taylor, Janet A. A personality scale of manifest anxiety. J. abnorm. soc. Psychol., 1953, h8, 285-290- Titus, H. E., and Hollander, E. P. The California F. Scale in psychological research., Psychol. Bull., 1957, 5h, h7-6h. Varela, Jacobo A. A cross—cultural replication of an experiment invol- ving birth order. Wpsoc. Psychol. , 196,4, 69, 1756-1457. Vidulich, R. N., and Kaiman, I. P. The effect of information source status and dogmatism.on conformity behavior. J. abnorm. soc. Psychol., 1961, 63, 639—6h2. warren, J. R. Birth order and social behavior. Psychol. Bull., 1966, 65 , 38-149 0 weller, L. The relationship of birth order to anxiety. Sociometry, 1962. 25. hlS-hl7. Welsh, G. S. An anxiety index and an internalization ratio for the MMPI. J. consult. Ps chol., 1952, 16, 65-72. Welsh, G. S. Factor dimensions A_and R, In G. S. welsh and W. G. Dahlstrom (Eds.), Basic readings on the MMPI in psychology and medicine. MHnnespolis: University of Minnesota Press, 1956. Pp. 265-281. White, R. W. The abnormal personality (2nd ed.). New York: Ronald, 1956. wrightsman, L. Effects of waiting with others on changes in level of felt anxiety. J. abnorm. soc. Psychol., 1960, 61, 216-222. Yaryan, Ruby 3., and Festinger, L. Preparatory action and belief in the probable occurence of future events. J. abnorm. soc. Psychol., 1961, 63, 603-606. Appendix A 87 DO NOT PUT YOUR NAME ON THE PAPER. Just fill in the following blanks. Present date Date of birth Age .. month day year Place of birth town state Grade in school a Male Female Religion 1“ Are you: ........... an only child? ,,,,,,,,,,, a fifth-born child? "M” a first-born child? ,,,,,,,,,,,,, a sixth-born child? .............. , a second-born child? .11, a seventh-born child? .aam- a third-born Child? .WW- an eighth or higher- born child? -- a fourth-born child? 0n the following pages you will find a list of to statements. These are what many people think and feel about a number of important issues concerning both ourselves and the world.we live in. The best answer to each statement is your personal Opinion, your own point of view. You may find yourself agreeing strangly with some of the statements, disagreeing just as strongly with others, and perhaps not very sure about others. Whether you AGREE or DISAGREE with any statement, you can be sure that many other people feel the same as you do. Because these statements have been made by adults, some of them are hard for high school students to understand. We have asked another group of high school students to help us change the words in order to make the meanings clearer. They changed 25 of the ’40 statements. We have put the changes they made [in brackets] under the statements made by the adults. When you check whether you AGREE or DISAGREE 'with the statements you may understand the version [in brackets] better, but the two statements mean the same thing. The results of this study will be valuable in terms of understanding what high school students think and feel about the issues. While the identity of the individual student is not important to the study, the over-all results are of great importance. And so we would appreciate your careful, accurate, thoughtful, and conscientious cooperation in answering the questions. * it -* * Here are two sample statements. Note that you check your AGREEMENT on the left side, and your DISAGREEMENT on the right side. .Make only ONE check.per statement. I AGREE A. Fair play is not important I DISAGREE “mavery much in sports mg“ very much ..... on the whole mir.on the whole 111a little mm--a little I AGREE B. Some people think about them- I DISAGREE ”a, very much selves too much. ,,,,,,,,, very much .11.0n the Wh01€ “W“ on the whole “waa little , ,,,,,,, a little -2- REMEMBER: THIS IS NOT A TEST. We would like to see whether you agree of disagree with statements made by adults every day. PLEASE BE SURE TO MARK EVERY STATEMENT ONLY ONE CHECK PER STATEMENT * * * * * * I I AGREE The United States and Russia have just about :I DISAGREE -1. very much nothing in common. mm,very much ,1” on the whole mm,on the whole ,“ a little _____ a little I AGREE The highest form of government is a democracy, I DISAGREE , very much and the highest form of democracy is a govern- MM.Very much “a on the whole ment run by those who are most intelligent. .Mm.on the whole aw,a little _m_a little [The best kind of government is a democracy, and the best kind of democracy is a govern- ment run by those who are smartest.] I AGREE Even though freedom of speech for all groups I DISAGREE _* very much is a worthwhile goal, it is unfortunately -1 very much mm on the whole necessary to restrict the freedom of certain _ _____ on the whole “m a little political groups. mm a little [All groups should be able to say what they believe in without anyone stopping them, but some political groups have to be limited in this freedom, although it is too bad to have to limit them.] I AGREE It is only natural that a person would have I DISAGREE .11 very much a much better acquaintance with ideas he .flm,very much new on the whole believes in than with ideas he opposes. mg_on the whole - a little Wm.a little [It is only natural that a person would know much more about ideas he believes in than ideas he doesn't like at all.] I AGREE Man on his own is a helpless and miserable I DISAGREE mmvvery much creature. Wa.very much _w,on the whole WW.on the whole Nana little _MM a little ....-,_.1. ..- .1 ..-. - ..._j - 1..- -__._-_ __ I AGREE Fundamentally, the world we live in is a ;I DISAGREE ,,,,,,, very much pretty lonesome place. m“,very much ______ on the whole “much the whole m a little [Actually, the world We live in is a pretty m-_a little lonesome place.] ... ..‘t I I]! l ‘ I.‘ {III .- .I.I.\.‘. I! -3- *a- H. ‘<-- . I AGREE 7. Most people just don't give a "damn" for a I DISAGREE ...... very much others . l 1...... very much .1- on the whole .1- on the whole ,_m,a little ._. a little ...... .....- .....- ......i-._._.....--___..____...-.-,..... ......1... . ..._.._.. ! ”.....- .-. I AGREE 8. I'd like it if I could find someone who I I DISAGREE ,gw very much would tell me how to solve my personal ,,,,, very much .1. on the whole problems. _mgmon the whole ..1 a little "“"_a little . ...... ...... .....-“ -- .... ---.-.--. -....L.-._-._... ...... ......in -..—...“... .. I AGREE 9. It is only natural for a person to be I DISAGREE “_m very much rather fearful of the future. .11 very much _m“ on the whole ,g_ on the whole “mm a little ”WW a little I AGREE 10. There is so much to be done and so I DISAGREE mmm.very much little time to do it in. mw_ very much “mm on the whole ,g”. on the whole ,mm a little .1. a little I AGREE 11. Once I get wound up in a heated dis- I DISAGREE a.” very much cussion, I just can't stop. _g_ very much .1“ on the whole m“. on the whole “in a little “w” a little I AGREE ‘ 12. In a discussion, I often find it neces- I DISAGREE "“wmvery much sary to repeat myself several times to ..... Very much “a on the whole make sure I'm.being understood. ,mmm on the whole 11m.a little .mm a little [In a discussion, I often have to say the same thing several times to make sure others understand me.) I AGREE 13. In a heated discussion I generally be- I DISAGREE mm“ very much come so absorbed in what I am.going to ,,,,,,,,, , very much .m on the whole say that I forget to listen to what ,mMnon the whole ,,~, a little the others are saying. ,me.a little 3 [In a heated discussion, I almost always * think so hard about what I am going to ' say that I forget to listen to what the 2 others are saying.) -l-nmmn_1_1uneluwr.___,l11- - .1_.l. .__W.N_T-i_,.-,n_l”“Mlu, I AGREE ilk. It is better to be a dead hero than to I DISAGREE "mm,very'much f be a live coward. ”.1.very much w_m on the whole é mm“ on the whole .11 a little n., a little 11-_______11-_1-f._1l .111 __ a .--1._ I AGREE {15. While I don't like to admit this even to I DISAGREE mnmwvery much i myself, my secret ambition is to become -w",very'much www.0n the whole 3 a great man, like Einstein, or Beethoven, “m,on the whole WM,a little or Shakespeare. - a little .......... [While I don't like to say this even to myself, my secret goal in life is to become a very great man.] -h- I AGREE 216. The main thing in life is for a person to ’ I DISAGREE mm, very much 3 want to do something important. .1” very much ........ on the whole 3 ‘ -_. on the whole in“ a little i I in” a little I AGREE 117. If given the chance, I would do something I DISAGREE _MWW very much E of great benefit to the world. ,“m,very'much in“ on the whole a mmm_on the whole .nr- a little 3 [If given the chance, I would do some— ,gm a little } thing of great help to the world.) , 11 I AGREE ‘18. In the history of mankind there have I DISAGREE - A very much probably been just a handful of great in“.very'much Vernon the whole thinkers. ~m__on the whole “a“ a little ”in a little I AGREE 19. There are a number of people I have come I DISAGREE mmm,very much to hate because of the things they stand mmm,very'much .Mm.on the whole ; for. “Mm on the whole all a little i ..“1 a little I AGREE 20. A man who does not believe in some great I DISAGREE -11 very much , cause has not really lived. mm“,very'much in“ on the whole ? mmw_on the whole -11 a little [A man who does not believe in something “mula little , which can be of great help to the world L has not really lived.] .1._1__1n-1__. 1,- ___ I AGREE ’21. It is only when a person devotes himself I DISAGREE u very much 3 to an ideal or cause that life becomes ,,,,,,, very much Man on the whole i meaningful. .nwn.on the whole _mw- a little 3 mam.a little 5 [Life takes on rea1.meaning for a person I only when he puts all of his effort into 3 something which has a high purpose or ; which is important in this world.) I AGREE 322. Of all the different philosophies which I DISAGREE .1“ very much é exist in this world, there is probably “muhvery'much ”um on the whole ? only one which is correct. awn” on the whole -. a little ! .mww.a little ; [Probably only one of the different beliefs ’ a which there are in this world is right.) J I AGREE €23. A person who gets enthusiastic about too , I DISAGREE ,1 very much f many causes is likely to be a pretty .-. very much' ,,,,, on the whole 3 "wishy-washy" sort of person. mg” on the whole- un.a little .W, a little [A person who gets interested and excited about too many things which are important in this world is likely to be someone who can't make up his mind.] -5- I AGREE [2A. To compromise with our political opponents 3 I DISAGREE ”an very much 3 is dangerous because it usually leads to ,,,,,,,, very much ...... on the whole j the betrayal of our own side. -- on the whole ......... a little 5 -1, a little 3 [It is dangerous to give in, even a little, ' to people on the other side of the political fence, because this usually leads to the 7 weakening or defeat of our own side.] u----_mmni.nl ._ it u. ______,..n_1__. -n“1._.____.m____. I AGREE 25. When it comes to differences of opinion in I DISAGREE m . very much religion we must be careful not to compro- ,mm very much ,WW on the whole mise with those who believe differently __g on the whole _, a little from the way we do. ”_W a little [When it comes to differences of opinion in religion, we must be careful not to give and take with those whose beliefs are dif- ferent from ours.] I AGREE 26. In times like these, a person must be pretty I DISAGREE ......... very much selfish if he considers promarily his own “. very much ”w, on the whole happiness. _____ on the whole ..... a little . an” a little ; [In times like these, a person must be 1 pretty selfish if he puts his own happi- -i ness ahead of anything else.) I AGREE ‘27. The worst crime a person could commit is I DISAGREE ,,,,,,, very much E to attack publicly the people who believe wm very much 1.” on the whole i in the same thing he does. ...... on the whole _ - a little 3 ,,,,,,, a little ’ [The very worst thing a person could do I is to openly attack the people who believe i in the same thing he does.] __n_m«_a_im.-_l,~-,____“n__,”_._._““1_._u_“w.ww._-._.u_m".1,um*_,_w,,.mwu_1n_.-.. I AGREE £28. In times like these it is often necessary I DISAGREE ,,,,,,, very much I to be more on guard against ideas put out a”. very much W. on the whole 3 by peOple or groups in one's own camp than m,” on the whole mm a little E by those in the opposing camp. ,,,,,,,, a little 1 i [In times like these, we often have to be 5 more careful about dangerous ideas put out 7 by people or groups on our side of the , g fence than by people on the other side of f the fence.] -am_w..a-_“e..___ma.-1_,-_1_”mli.r~wu-is .--“.MU.--“11_._w.wm.11,. I AGREE .29. A group which tolerates too much difference I DISAGREE a- very much ; of Opinion among its own members cannot ..... very much _m“ on the whole ? exist for long. * ,,,,,,, on the whole , a little ? ,,,,,,, a little i [A group which allows too much difference of opinion among its own members cannot last long.] lollllllnl.‘ -6- I AGREE 30. There are two kinds of peeple in this world: I DISAGREE ,.rgvery much those who are for the truth and those who in“ very much an“ on the whole i are against the truth. ”A- on the whole in” a little ,,,,,,, a little I AGREE Q31. My blood boils whenever a person stubbornly I DISAGREE fl”, very much ; refuses to admit he's wrong. “gm very much ”mm on the whole 3 -. on the whole -- a little [I get very angry when a person just won't ”mm a little ‘ admit he's wrong.) I AGREE {32. A person who thinks primarily of his own I DISAGREE ,,,,,,, very much I happiness is beneath contempt. mmjvery'much “WW on the whole ; ‘ .mm-0n the whole .m a little 3 [A person who thinks first of his own ..... a little happiness is about as low as anyone can ' get.) --Wlmllm-l.--lli ll ..___lll.l_.___l_ll_fll.1l-.... ........ .11 I AGREE 333. .Most of the ideas which get printed nowadays I DISAGREE “W very much § aren't worth the paper they are printed on. _mm very much an“ on the whole 7 -M on the whole .3”- a little f ,WW a little --..--“ ”all-.- _- 1.....-.--_..-..-....._... M--. .....- ---ll-._-....-...l ..." “...-..” -.. .. Annual- I AGREE E3u. In this complicated world of ours, the I DISAGREE - very much only way we can know what's going on is in. very much .- on the whole ; to rely on leaders or experts who can .mm on the whole -. a little 3 be trusted. ..MA a little i [Since this world of ours is so hard to ‘ understand, the only way we can know i what's going on is to depend on the § leaders and those who know a lot, whom ; we can trust.) I AGREE £35. It is often desirable to reserve judgement I DISAGREE A.” very much ‘ about what's going on until one has had a -1- very much “mm on the whole ; chance to hear the opinions of those one ....... on the whole ,,,,,,, a little ; respects. "um a little V [It is often better to wait until people . you think a lot of have given their opinion g before you make up your own mind about 2 what's going on. J I I AGREE €36. In the long run, the best way to live is I DISAGREE ”Wk very much f to pick friends and associates whose ,mg very much ,,,,,,,,, on the whole 5 tastes and beliefs are the same as one's .1“ on the whole ._l. a little i own. ._".a little . ...... 1 ...... .. --- ... l - I AGREE T3 The present is all to often full of un- I DISAGREE .1“ very much ; happiness. It is only the future that 1 ...... very much _ .11 on the whole 1 counts. mpg on the whole ,MMra little ‘ _mm_a little -7- l I AGREE i 38. If a man is to accomplish his mission in i I DISAGREE ._” very much 7 life it is sometimes necessary to gamble ‘ mum very much up on the whole) "all or nothing at all." i “n“ on the whole _______ a little ' l a little I [If a person is to reach his goal in life, 3 ' he sometimes has to pla an "all" or 7 "nothing" kind of game.) ‘ I AGREE 39. Unfortunately, a good many people with I DISAGREE "gm very much whom I have discussed important social .ll very much .I on the whole and moral problems don‘t really under- an- on the whole in, a little stand what's going on. ,,,,,,,,,, a little [I think it's too bad, but lots of people . I've talked to don‘t really understand I ; about the important problems facing the 9 ' world today or about what is right and 2 what is wrong.) ; I AGREE #0. Most people just don't know'what's good :1 DISAGREE m very much for them. : all very much .l. on the whole 3mm, on the whole - a little “W,a little .1.“ -... .nu. . . .. .... "In" In - u w “- ' ... STOP. Please go back and make certain that: “a... ...-... .‘u ...... n a _. .x A... ..u. I- l ...-- "m- . , w v a) you have answered each and every one of the ho statements, and b) you have made only one check for each of the ho statements. Thank you for your cooperation. Appendix B V 95 DO NOT PUT YOUR NAME ON THE PAPER. Present date "1-, Just fill in the following blanks. Date of birth AgeinMlle month day year (not 1965) Place of birth town state Psychology 151 section Malegwwmmmm Female "lawman”. Religion Are you: gm ”flan only child? .Nwllm.a fifth-born child? a first-born (oldest) child? -s mamas-ul- “mall,a second-born child? .m~mmmna third-born child? _mmwwm a fourth-born child? c a snowman-”DO- Us .a sixth-born child? - ‘QIC‘OOOso s a... no. a seventh- or later- born child? unnamma twin? Anal" a triplet? If you are one of several children, please list the exact ages of your brothers and sisters. Brothers 1- ........ years and ...... months old. 2- "alwyears and W“.months old. 3._meuyears and_,w,months old. h. mmmflyears and ”“_months old. Sisters 1.,Mm” years and.mg months old. 2..M,"-years and M,_months old. 3. mmmmyears and “m months old. no.1AMIyears and mwmmonths old. How nervous or uneasy do you feel about taking part in this experiment? ‘ Please answer by checking one of the following alternatives: _,I feel extremely uneasy. o-OOOQOOoooa.” __I feel very uneasy. “Ow-”1 a... ,._ln.I feel quite uneasy. shall I feel a little uneasy. m I feel relatively calm. I ”Own. 1.1.“ I feel completely calm, Below are 3h statements concerning a.number of important issues about our- selves and the world we live in. These are all controversial statements, and.many conflicting and opposing points of view are included. You may find yourself agree- ing strongly with some of the statements, disagreeing just as strongly with others, and perhaps not very sure about yet others. Whether you agree or disagree with any statement, you can be sure that there are many other people who feel pretty much the way you do. So the best answer to each statement is ygur personal opinion, your own point of view. Please try to respond to all the statements as honestly and frankly as you can. No one will know how you respond because we are asking you not to identify yourself. Instructions: Respond to each of the statements using the following numbers, depending on how you feel in each case: +3: I AGREE VERY MUCH -3: I DISAGREE VERY MUCH +2: I AGREE ON THE WHOLE -2: I DISAGREE ON THE WHOLE +1: I AGREE A LITTLE -l: I DISAGREE A LITTLE For example, if you agree very much with a statement, write +3 on the short line preceeding that statement. If you happen to disagree a little, write -1 in front of it. Respond to each statement as best you can. Please do not omit any. Go rapidly but carefully. Do not spend too much time on any one statement; try to respond and then go on. Do not sign your name. REMEMBER: PLEASE BE SURE TO ANSWER EVERY STATEMENT. +3: +2: +18 "1...--.” 2 . woo-usuooanov I no. 0 unto-000.00.00.00... - 9 lO. 0. h anmn-acos'ao ll. . llfl‘fl“.‘-Coflnin 12. I on w.- 000-..... -.....m....... 13 - W... 15 - 11.11.16. I AGREE VERY MUCH -3: I DISAGREE VERY MUCH I AGREE ON THE WHOLE -2: I DISAGREE ON THE WHOLE I AGREE A LITTLE -l: I DISAGREE A LITTLE If given the chance I would do something of great benefit to the world. Nowadays when so many different kinds of people move around and mix together so much, a person has to protect himself egpecially carefully against catching an infection or disease from.themt It is often desireable to reserve judgement about what's going on until one has had a chance to hear the opinions of those one respects. Homosexuals are hardly better than criminals and ought to be severely punished. In a heated discussion I generally become so absorbed in what I am going to say that I forget to listen to what the others are saying. A person who gets enthusiastic about too many causes is likely to be a "wishy-washy" sort of person. Young people sometimes get rebellious ideas, but as they grow up they ought to get over them and settle down. In the history of mankind there have probably been just a handful of really great thinkers. Some day it will probably be shown that astrology can explain a lot of things. The worst crime a person could commit is to attack publicly the people who believe in the same thing he does. There is hardly anything lower than a person who does not feel a great love, gratitude, and respect for his parents. Sex crimes, such as rape and attacks on children, deserve more than mere imprisonment; such criminals ought to be publicly whipped, or worse. People can be divided into two distinct classes: the weak and the strong. The wild sex life of the old Greeks and Romans was tame compared to some of the goings-on in this country, even in places where people might least expect it. Unfortunately, a good many people with whom I have discussed important social and moral problems don't really understand what's going on. What the youth needs most is strict discipline, rugged determination, and the will to work and fight for family and country. ...-06”“...- a o . ”*OO“~*OO~I I.“ so- 0“. s a... - nun-Ion. “ 4‘0 coo-"mo new. coon-om... .- uowhuuo‘u - . soul‘s-Dusscjo-..‘ no.“ --.-...”... 17. 18. 19. 20. 21. 22. 23. 2h. , 25. 26. 27. 28. , 29. . 30- 31. 32. - 33- _ 3h. B The highest form of government is a democracy and the highest form of decmocracy is a government run by those who are the most intelligent. In times like these, a person.must be pretty selfish if he considers primarily his own happiness. While I don't like to admit this even to myself, my secret ambition is to become a great man, like Einstein, or Beethoven, or Shakespeare. No weakness or difficulty can hold us back if we have enough will power. Most pe0ple don't realize how much our lives are controlled by plots hatched in secret places. Of all the different philosophies which exist in this world there is probably only one which is correct. Even though freedom of speech for all groups is a worthwhile goal, it is unfortunately necessary to restrict the freedom of certain political groups. Most people just don't know what's good for theme There are two kinds of people in this world: those who are for the truth and those who are against the truth. Fundamentally, the world we live in is a pretty lonesome place. It is essential for learning or effective work that our teachers or bosses outline in detail what is to be done and exactly how to go about it. Wars and social troubles may someday be ended by an earthquake or flood that will destroy the whole world. A person who thinks primarily of his own happiness is beneath contempt. If people would talk less and work more, everybody would be better off. It is only when a person devotes himself to an ideal or cause that life becomes meaningful. If a man is to accomplish his mission in life it is sometimes necessary to gamble "all or nothing at all." In a discussion I often find it necessary to repeat myself several times to make sure I am.being understood. The present is all too often full of unhappiness. It is only the future that counts. hwy—— Please go back over your paper and make certain that you have answered each and every one of the statements. How would you rate your general state of health? Please place a check mark over the appropriate number. Very good Very poor '0’“ I“... wwwuvtmo' Oho... “0.00:. no..." “bl-mom oooooooooooooooooooo om lllllllll . ...- ooooooooooooooo - moo- IOCOI'IIIC lull. .....oO-u nnnnnn no "can“ unnu- -0.“ no..." on . .- +3 +2: +1: 12. 13. 1h. I AGREE VERY MUCH -3: I DISAGREE VERY MUCH I AGREE ON TEE WHOLE -2: I DISAGREE ON THE WHOLE I AGREE A LITTLE -l: I DISAGREE A LITTLE There is so much to be done and so little time to do it in. Every person should have complete faith in some supernatural power whose decisions he Obeys without question. The businessman and the manufacturer are much more important to society than the artist and the professor. Familiarity breeds contempt. The United States and Russia have just about nothing in common. Most of our social problems would be solved if we could somehow get rid of the immoral, crooked, and feebleqminded people. What this country needs most, more than laws and political programs, is a few courageous, tireless, devoted leaders in whom the people can put their faith. It is only natural that a person should have a much better acquaintance with ideas he believes in than with ideas he opposes. My blood boils whenever a person stubbornly refuses to admit he's wrong. There are a number of persons I have come to hate because of the things they stand for. When a person has a problem.or worry, it is best for him.not to think about it, but to keep busy with more cheerful things. No sane, normal, decent person could ever think of hurting a close friend or relative. A person who has bad manners, habits, and breeding can hardly expect to get along with decent people. A man who does not believe in some great cause has not really lived. The main thing in life is for a person to want to do something important. Once I get wound up in a heated discussion I just can't stOp. Science has its place, but there are many important things that can never be understood by the hwman mind. Some people are born with the urge to jump from high places. In this complicated world of ours the only way we can know what is going ion is to rely on leaders or experts who can be trusted. ammowowo'ur . Woo-.- can..." u no to - I.- .00... .’ ,iwmmm.25o .26. ............ 27- 28. aouomo-ooatnodl .................. 29- .na Inn- on- ooooo I'd like it if I could find someone who would tell me how to solve my personal problems. A group which tolerates too much differences of opinion among its own members cannot exist for long. When it comes to differences of opinion in religion we must be careful not to compromise with those who believe differently from the way we do. Nobody ever learned anything really important except through suffering. In the long run the best way to live is to pick friends and associates whose tastes and beliefs are the same as one's own. Most of the ideas which get printed nowadays aren't worth the paper they are printed on. Nowadays more and more people are prying into matters that should remain personal and private. It is only natural for a person to be rather fearful of the future. Obedience and respect for authority are the two most important virtues children should learn. In times like these it is often necessary to be more on guard against ideas put out by people or groups in one's own camp than by those in the opposing camp. Most people just don't give a "damn" for others. Man on his own is a helpless and.miserable creature. To compromise with our political opponents is dangerous because it usually leads to betrayal of our own side. An insult to our honor should always be punished. It is better to be a dead hero than a live coward. Please go back over your paper and make certain that you have answered each and every one of the statements. How nervous or uneasy do you feel about taking part in this experiment? Please answer by checking one of the following alternatives: ”.1 feel extremely uneasy. - ...... .0. * .mMH“.I feel very uneasy. ,,,,,,,,,,,,,, I feel quite uneasy. WWWMW.I feel a little uneasy. MWMMM I feel relatively calm. _Mwmmm I feel completely calm” Would you like to leave rather than proceed with the experiment? Yes No M. .................. - ......l . o Imowuo‘ac 0M- "‘c‘ffllllflllfilfllllllflr