gm EWEém mamas M29 m mwaegsssmm mmmswga megahrfiwmémaféa.k sue-imam flaw mas ifiaéww fiaaamhuzg €933 ' E 4’— M!!!" o "“ -r-— .. Q i 1 f Thlsiotoeertifgthntthe 3“; thesis entitled . n'.[n1.'.erv:i.ewer Problems and the Non-Professional Intervieweru . , presented bl] Helenan Sonnenburg has been accepted towards fulfillment of the requirements for 11.5. degree mm 8; Anthropology C: ‘ ' Major professor t Date June 2- 1-253 t 0-169 u-_—- ‘WJ‘ -£;‘_-L.- in -. '— L - . '-..... .4...“ I ‘--'-441.—.4—-—- . Lax- -..._ . - ,Lfi‘ av . . I Vll x \x int ..n\ VI (.1 . \\ \ II ; . ..\.Ia. . . . . I. ..\ . . lol -¢ . (. r . . VK u.‘/. y . v 5‘ b\ .u. k . . d .r I If All . tr. .. VII}! 4 .oo J. I. \ v , / o s A . .... .n . 1. .1 . I: e \ .o ' . S. . .. .5... . . A’...’ 1‘ .‘ o It -. . I». at \‘H . . “1 .,,H .p 5 .W: t c d.. .9. . . r T. x “.V C .. . » ‘ . ft .. 7 _ u ‘ Hr. .... . .0 1.57"" . ...L.wny3.§m~&i.fil. “114%! «1341111. . .yfi l n ‘J a snuflvll, u was“; . 1.7.... 109..“th 4W4 ,3 , fl « all... . .. , . .. u in: iuua'11fl, 5 Dr I x .. ~ . . 7.. . u t . | HI. h. lll .H‘ .. . r}. €6.th I’ll .LFl K.» . . . .hv-.l,«w.h Ms. {.91} i r l Ir 'Hl'r'»*: INTERVIEWER PROBLEMS,AND THE NON-PROFESSIONAL INTERVIEWER By Helenan Sonnenburg A THESIS Submitted to the School of Graduate Studies of Michigan State College of’Agriculture and.Applied Science in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology and AnthrOpology 1953 'IHESIS (0 /L/§'3 ACKNOWLEDGMENTS This study has made use of data collected as part of a research project Sponsored by the Michigan State College Social Research Service of the Department of Sociology and Anthropology. It was financed by a grant from the Health Information Foundation of New York City. Over- all supervision of this research has been under the direction of Dr. Charles P. Loomis, Director of the Social Research Service, and Dr. Charles R. Hoffer, Acting director in the absence of Dr. Loomis. Dr. Christopher Sewer and Dr. John B. Holland were corchairmen of the research. The writer wishes to express sincerest gratitude to Dr. Christopher Sower for his time and effort eXpended in the supervision of this thesis problem; to Dr. John B. Holland for his assistance in deve10ping the research design; to Dr. Duane L. Gibson for his suggestions concerning research.methodology; to Dr. Charles R. Hoffer for his assistance with the preparation of the thesis. There are other members of the project committee to whom the writer would like to express appreciation for their friendship and co- Operation throughout the research. These are: Dr. John F. Thaden, Dr.’Hilbur Brookover, Dr. J; Allan.Beegle, Professor Paul.Miller, Professor David Steinicke, and Mr; Halter Freeman. Finally, the writer would like to thank Dr. Harry Scales and Dr. George Belknap for encouraging the completion of this thesis and for expressing confidence in the writer's ability to continue graduate study. 63w. «r m n r a)“; _i‘ «gag CHAPTER .II III IV TABLE OF CONTENTS PAGE mTRODUCTIONOO......OCOOOO ..... 00.00..........OOOOOOOOOOO l A. Lenawee County Health and Defense Self-Survey. ..... 2 B. Statement Of the HoblemOOOO.......OOOOOOOOOOOOOOOO 6 C. Clarification of the Terms Used in the Study....... 19 REVIEW OF LIil‘m‘TUR-EOOOOO0.00.00.00.00... oooooo Oooooooooo 21 A. Interviewer Problems.......................... ..... 21 B. Qualifications for Interviewers.................... 2h C. Previous Experiences of the Interviewers........... 30 D. Interviewer Training......................... ..... . 32 E. Community Self-SurveyS............................. 3h FmTHOmIDGYOOOOOOOOO... ........ OOOOOOOOOOOOOOOOO0.0000... ho A Interview Schedule........... ........... ... ...... . ho B. Testing of the Interview Schedule.................. h3 C. The Poplfl'ation. O.......OOOOOOOOOOOOOOIOOO0.00...... lib D. The Interviewers............................ ...... . Lb E. Developing the Code and.Ana1ysing the Data,.,,,.,.. LS F. Chi-Square Method of Analysis............. ...... 1:5 PROBLhMS REPORTED BY THE INTERVIEJEBS....... ........ h? A. The Health and.Defense Questionnaire............... h? B. Types of Problems Reported by the Interviewers..... Sh C. Limitations of This Method for Defining Interviewer PrOblemSOOOO....00............OOOOOOOOOO0.0.0....0O 57 TESTS FOR VARIABLES mama TO mrmvrmm PROBLEMS ...... S9 A. Social Characteristics of the Interviewers......... 59 B. Previous Experiences of the Interviewers ....... .... 68 C. Involvement and Participation of the Interviewers in the Survey.......... ...... . ......... 78 TRAINING OF THE INTERVIEWERS.. ........... . ........... .... 89 A Types of Interviewer Training...................... 89 B. Opinions About the Sufficiency of the Training..... 92 C. Types of Training and Reported Interviewer Problems 95 D. ASpects of the Training Pragram.................... lOO TABLE Or“ CONTENTS - Continued CHAPTER PAGE VII CONCLUSIONS... ...... . ..... . ...... .106 BIBLIOCfiAPhY... ....... ....... 111 TABLE III IV VI VII VIII XI XII XIII XIV LIST OF TABLES PAGE Distribution of the Interviewers Reporting Problems on the Civil Defense Questions by the Areas on Which They Reported Pro‘blemso0000.00000-oooo-orooooooooooooooo00000000000000... Distribution of the Interviewers Reporting Problems on the Health Questions by the Areas on.Which.They Rad Problems... Distribution of the Interviewers by the Types of Problems Which They Reported on the Entire Survey Questionnaire . . . . . Distribution of the Types of Interviewer Problems by the Sex of the InterviewerS.................................... Distribution of the Interviewers by Age.................... Distribution of the Types of Interviewer Problems by the .Age of the Interviewers..... Distribution of the Interviewers by Amount of Education.... Distribution of the Types of Interviewer Problems by the Education of the InterviewerS.............................. Distribution of the Types of Interviewer Problems by the Location of the Interviewer's Residence in.Lenawee County.. Distribution of the Types of Interviewer Problems by the Previous Interviewing Experience of the Interviewers....... Distribution of the Types of Interviewer Problems by the Previous Door-to-Door Sales Experience of the Interviewers. Distribution of the Types of Interviewer Problems by the Previous Employment Experience of the Interviewers......... Distribution of the Types of Interviewer Problems by Professional Experience.................................... Distribution of Interviewers by Organizational Affiliation. Distribution of the Types of Interviewer Problems by the Number of Organizational.Affiliations...................... SO Sh 57 62 63 6h 65 67 69 71 73 7h 75 76 LIST OF TABLES - Continued TABLE XVI XVII XVIII XIX XXI XXII XXIII XXIV XXVI XXVII XXVIII PAGE Distribution of the Types of Interviewer Problems by Previous Participation in Community Activities.. .......... . 77 Distribution of the Types of Interviewer Problems by the Method of Township Organization............................ 80 Distribution of the Types of Interviewer Problems by the Structural Position of the Interviewer in the Survey organizationOOOOOOOO0.00.0000........OCOOOOOOOOOIOOIOOO0.0. 81 Distribution of the Types of Interviewer Problems by Inter- Viewer Training. 0 O O C O O O O C O O. O O O O I O O O O C O O O O. O O O O. O O O O O O O O O O 0 83 Distribution of the Types of Interviewer Problems by the Number of Interviews Completed by the Interviewer.......... 8h Distribution of the Types of Interviewer Problems by the Extent to Which the Informants Had Heard About the Survey Before Being Contacted for an Interview.................... 86 Distribution of the Interviewers by Their Opinions.About the Sufficiency of Their Training.......................... 93 Distribution of the Types of Interviewer Problems by the Opinions of the Interviewers as to the Sufficiency of Their Training.oooooooooooooooooooooo.ooooooooooooooooooooooooooo 9L1 Distribution of the Types of Interviewer Problems by Individually and Non-Trained Interviewers......... ....... .. 95 Distribution of the Types of Interviewer Problems by Individually and Group Trained Interviewers....... ........ . 9? Distribution of the Types of Interviewer Problems by the Types of Training.Meetings.Attended by the Interviewers.... 98 Distribution of the Types of Interviewer Problems by Train- ing Meetings Not Attended by Michigan State College Personnel-......OOOOO......OOOCOOOOCOOOOOOOO.00.0.0.0...O.O. 99 Distribution of the Types of Interviewer Problems by the Interviewers Who Have and Have Not Read the Instructions... lOl Distribution of the Types of Interviewer Problems by the Interviewers Who Have and Have Not Experienced a Practice InterviewOOOOOOIoOOOOOOOOOOIOOO ..... OOOOOOOOOOOOOOOOOOIOOCO102 LIST Oi" TABLES - Continued TABLE PR GE XXX Distribution of the Types of Interviewer Training by the Interviewers Who Have and Have Not Read the Instructions. .. 103 XXXI Distribution of the Types of Interviewer Training by the Interviewers Who Have and Have Not Experienced a Practice IntervieWOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 103 CHAPTER I INTRODUCTION Public Opinion and other research agencies are continually con- cerned with the techniques they employ to obtain information. Particularly have they been interested in.improving the quality of the interviewing done during professional surveys. Attempts have been.made to measure the quality of the interviewing according to the campleteness of the schedules and according to certain other defined criteria of quality; Concern has been paid to interviewer bias as it affects the character and quality of the interviewing and to certain problems which affect the cost of the fact-finding, such as the ncheater problem." In each case, some consideration has been given to the characteristics of the interviewers which are related to the quality of their interviewb ing. Little systematic attention, however, has been.paid to the problems which interviewers encounter and the effect which these have on the quality of the interviewing. For this reason, this thesis is concerned with the reported problems of a group of noneprofessional, volunteer interviewers who participated in a self-survey of health and defense.1 l. A self-survey, as distinguished from other social surveys, is a method of fact-finding which is planned, organized, and carried out by the citizens of the county or community in which it takes place, rather than by professional survey agencies or specialists. The study will be directed toward discovering the factors related to the problems reported by the interviewers. It is not assumed that the find- ings will necessarily hold true for professional, paid interviewers who participate in the public Opinion and attitude surveys of agencies, such as the National Opinion Research Center. However, this study may suggest a method which could be employed in evaluating the factors re- lated to paid, professional interviewing. A description of the self-survey carried out in Lenawee County, Michigan, a statement of the thesis problem and the hypotheses to be tested, and a clarification of some of the terms employed, will follows A. Lenawee County Health and Defense Self-Survey During 1951, a self-survey of health and defense was carried out in the rural areas of Lenawee County,iMichigan.2 The Lenawee County Health Council, with the cOOperation of the County Health Department, the County'Extension Office, the County School Office, the County-Medical Society, and other organizations interested in health, sponsored the survey. The Health Specialists of the Michigan State College COOperative Extension Service served as consultant for the survey. These organiza- tions Sponsored it to groups in the rural townships and villages of the County and assisted them in organizing to conduct the health and defense survey in their local area. Assured that the survey was needed in 2. The survey was later carried out in the two urban areas, Tecumseh and Adrian, in 1952. The information concerning these surveys was not available for analysis at the time that this study was started. Lenawee County, local groups organized for the survey in one of three ways :3 1. In some of the local units, one or more individuals agreed to be chairmen.and to contact their friends, acquaintances, and people whom they did not even know, to do the interviewing. Organization is said to have proceeded there along friendship lines. 2. In other of the townships and villages, members of a formal organization, such as the Child Study Club or the local Extension.Club, voted to make the survey a project of their organization. jHany of the members agreed to participate or to get a friend to help with the interviewing. Involvement is said to have proceeded along association lines. 3. Finally, in some of the local units, the school district supervisors pledged their support to the self-survey idea and agreed to take the reSponsibility for completion of the survey there. Organization is said to have proceeded along school district lines. The interviewers for the self-survey in all cases were volunteer, non- professional interviewers. Few’had had any previous experience in door— to-door fact-finding; none had received previous professional interviewer training. 3. A fourth.method of organization is suggested by the data concerning the survey in.Adrian. Involvement of the interviewers and organi- zation for the survey was carried.out along precinct lines. Throughout the planning, organizing, and conducting of the survey, the participants and events were studied from a research standpoint by the members of the Social Research Service of the Department of Sociology and AnthrOpolOgy at Michigan State College. The research design for the Lenawee County study was oriented toward a concern with the self-survey as a community action process. The project was designed to find out what social processes were involved in a county-wide self-survey and what type of social structure was temporarily erected in organizing for the survey. They also were concerned with analyzing how permanent organizations, such as the County Health.Department and the Cooperative Extension Service, and less permanent organizations, such as the Health Council, utilize temporary and/or subordinate organizations to initiate and carry out action programs. The following research.methods were utilized: formal and informal interviews with the organizers at the county level; attendance at and observation of County Health Council meetings; formal interviews and in- formal conversations with the organizers in the townships and villages; a dayabybday recording of the events associated with the survey by the Extension Health Specialist of the Michigan.State College Cooperative Extension Service, who was the consultant for the county self-survey committee. The formal interview schedule used in interviewing the organizers at both the County and local levels was likewise used to interview the volunteer participants in the townships and villages. This schedule, designed by the research team included questions about the involvement of the interviewers, their training, previous experiences, problems, and knowledge of health action in the County. The data for this thesis have been obtained from these formal schedules. The author of this study was a member of the project staff which studied the Lenawee County Health and Defense Survey. This involved the following functions: reviewing of the literature written.about self- surveys which have been completed in other parts of the United States; assisting in the construction of the interview schedule mentioned above; assisting in the construction of the code which was used in the analysis of the schedules; participating in interviewing the interviewers; super- vising the coding of the schedules. ‘ During this experience, the author became interested in the volunteer participants who actually did the door-to-door interviewing, and particu- larly in their ability to obtain accurate findings when they had not received professional interviewer training. The original plan of this thesis included a resurvey of a sample of the informants interviewed during the health self-survey for the purpose of comparing the accuracy and completeness of the interviewing done by volunteers with that of professional interviewers. Such a resurvey, however, was not financially' possible. During the analysis of the schedules used in interviewing the participants, it was found that a considerable amount of the informa- tion concerning the interviewer training, previous experiences, education and interviewer problems of the volunteers had been obtained. The plan for testing the accuracy and completeness of nonpprofessional interviewb ing was replaced by-a research design concerned with the problems reported by volunteer participants duringwa self-survey. B. Statement of the Prdblem This study is concerned with the problems reported by volunteer, nonsprofessional interviewers in a self-Survey of health and defense.h Specifically; this study is based on the reported problems of 309 volun- teer participants in the self-survey described above.5 The volunteer‘ interviewers stated that they encountered two types of problems: pgoblems of communication (the interviewer had difficulty explaining the purpose of the survey or the purposes of specific questions to the re- spondent) and pgoblems of gpposition (the informant either refused the interview or greatly Opposed the efforts of the interviewer). .An at- tempt will be made to find out if there are any characteristics of the participants or any previous experiences which are related to the types of problems which they reported. A survey’of the literature related to fact-finding by the survey method, particularly that which was prepared by professional fact-finding agencies, indicates that much.time and Space has been devoted to outlin- ing the qualifications necessary for ”good" interviewers. Certain age, sex, education, etc. requirements are specified. Careful attention is h. These problems were reported by the interviewers and therefore may not be a true picture of the problems which were actually encount- ered. The limitations of using reported problems of the interviews ers as a variable will be discussed in Chapter IV. 5. There were 326 volunteer participants in the self-survey of the rural areas of Lenawee County who were interviewed by the research team for this study. Thirteen of these were organizers and did not interview. Four of the volunteer interviewers were not asked or did not reapond to the questions concerning interviewer problems. given to the training which the interviewer should experience before he carry out his job. The requirements and descriptions of the training programs are outlined in the Review of Literature. (See Chapter II) These requirements and others which have been suggested by the author's experiences with the Lenawee County study will be tested statistically in this thesis for their relationship to the problems reported by the interviewers. The variables which will be used for the statistical analysis of the stated problem have been divided into three categories: 1. Social Characteristics of the Interviewers--The variables - rfi - —- ‘I‘fi :- “Ir-a ‘F . , . . \/ included in this category are sex, age, education, and location of the interviewer's residence. 2. Previous Experiences of the Interviewers--The variables included in this category are previous interviewing experi- ence, sales experience, occupational experience, organi- zational membership, and participation in community activi- ties. 3. Involvement and Participation of the Interviewer in the ‘§ppzpy§-The variables included in this category are the method of township organization for the survey, the inter- viewer's position in the local survey organization, the training of the interviewer for the survey, and the extent to which the informants had heard about the survey before they were interviewed. The bases for selecting these variables, and their formulation into hypotheses will be described below: Social Characteristics of the Volunteer Interviewers There are several variables which are used frequently by sociolo- gists, psychologists, and anthropologists in their research. These are called social characteristics. Specifically, this term refers to factors such as age, sex, education, location of residence, marital status, religious affiliation, occupation, organizational membership, etc. They are used as variables mainly because research has shown that they are behavior related and behavior determining qualities. Role theory, such.as that outlined by Ralph Linton, suggests that peeple of a certain.age and sex, religious affiliation, education, etc. can be expected to behave in relatively specific ways in a given situation.6 In reviewing the qualifications required of paid interviewers by public Opinion.polling agencies, it can be seen that there is little specific agreement as to the social characteristics necessary to be a "good" interviewer. However, same rather general requirements are accepted, and these form the basis for the hypotheses which follow. How these were derived can be seen in the case of age. There is rather general agreement in the literature that older peOple do not make as good interviewers as do younger people. Fifty years is considered the dividing line. Above that age, interviewers have been 6. Ralph.Linton, ”Concepts of’Role and Status”, Swanson, Newcomb, and Hartley, Readings in Social Psychology, (New'fork: Henry Holt and Company, 1952) p. 263-266. observed to have more difficulty in "getting around" in bad weather, in gaining access to people's homes, are more opinionated, less scientific, etc.7 Whether they encounter more difficulties in communi- cating with the reSpondent and in overcoming Opposition is not actually suggested in the literature. This research will test voluntary inter- viewers conception of their communication and apposition difficulties by age differences. A hypothesis to this effect follows: Interviewers who are fifty years or older report problems of communication and cppgsition more frequently than do those who ...—'1‘: a are less than fifty Ears old. Among the authors reviewing the qualifications necessary for success- ful interviewing, the general agreement is. that women make better inter- viewers than men. A statement which has particular reference to this study was made by Edith Witt. She says, "Womenhgainreasier ‘_ entrance to hgm‘e‘s, stores, businesses, and have less difficulty stopping peOple on the street."8 If this were true,‘women_ should be expected to encounter fewer problems of Opposition, for much of the Opposition experienced is based on the reSpondent's dislike of being interviewed at all and is not due to the nature of the survey. The assumption is, likewise, that the same would apply for communication difficulties, although there is no written evidence to support this. The following hypothesis will be tested: k 7. Emory S. Bogardus. The New Social Reseapgh? (Los Angeles: Jesse Ray Miller, 1926), p. 56. 8. Edith Witt. "A Symposium of Interviewing Problems; the San Francisco Program for Improving Standards" , International Journal of Opinion and Attitude Research, 3 (1919), p. I139. 10 Men encounter pppblems of communication and opposition more frequently than do women. There is general agreement within the literature that the amount of education needed to interview successfully depends more on the difficulty ofpthe assignment than upon anything else. The interview form used in the Lenawee County study was highly structured, demanding very little interpretation on the part of the interviewers. It is assumed that there wmuld be no significant relationship between the amount of the education an interviewer had received and his ability to obtain complete and accurate data in the use of that particular interview schedule. Stated as a hypothesis, the above reads: Tpers is no difference amopg_the educational level of interviewers ‘ ~J‘ ”5"" in the extent to which.they report problems of communication and w..." 1.... opposition. The literature makes no reference to other social characteristics which are considered prerequisites to being hired or selected as an interviewer. However, it is considered necessary from a research stand- point to hold constant one other social characteristic. This variable is the location of the interviewer's residence in Lenawee County; The aspect of location.which is of concern here is that of farm vs. small city. The questionnaire used in the Lenawee County survey included some questions which had particular reference to farms. These were the ques- tions about the testing of cows for Bangs Disease and TB, the prevalence of'mastitis, and the source of water supply and sewage diaposal. These questions were considered unnecessary by many of the informants who lived in the towns and small cities. For this reason, it is suggested that there may be a significant relationship between locations of inter- viewer's residence and reported problems. The following hypothesis will be tested concerning this: Interviewers from the small cities repprted.problems of communi- Jr cation and opposition.more frequently:than did those from the '1‘ , r/qu‘u -a 1' rural areas. The above hypotheses which are concerned with the social character- istics of the interviewers will be tested in Part A of Chapter V. Previous Experiences of the Interviewers Questions about the previous experiences of the interviewers in door-to-door selling, interviewing, working outside the home, partici- pating in community activities, etc. were included in the interview schedule on the assumption that such experiences might be related to the interviewer's ability to do the voluntary interviewing job. It was felt that the participants who had had.previous experience particularly in door-to-door fact-finding, fund collecting, or selling might have less difficulty in gaining rapport, in explaining the idea of the survey, and in overcoming Opposition. The hypotheses to be tested concerning these experiences can be stated as follows: Interviewers who have had previous interviewing experience a- «Mem’fl' '"‘ report problems of communication and opposition less fre- flat-v ‘ (Cbhxur‘WW-i gpent£y_than do those who have no such experience. _ ,11i. 12 Interviewers who have had_previous experience in door-to-door selling4r§port problems of communication and Opposition less frequently than do those who have had no such experience. That the interviewers have had previous occupational experience outside of the home is not required by any of the agencies employing interviewers, nor does there appear in the literature any mention of the value of occupational experience in interviewing. For this reason, it is hypothesized that previous employment is not a factor that is re- lated to interviewer problems: There is no difference between distributions of interviewers py previous occupational experience in the extent to which they report problems of communication and Opposition. Information about the interviewer's present participation in organi- zations in the community was requested with the purpose of determining the relationship between organizational membership and.participation in the health survey. It was found that only lhi of the interviewers were not members of at least one organization besides the Church. In this study, it is considered important to determine if there is a relationship between.organizational membership and interviewer problems. If a re- lationship is found to exist between previous interviewing experience and interviewer problems, it is expected that there may also be a re- lationship between organizational membership experiences and the types of problems reported dependent to some degree on the extent to which the individual actually participated in the program of his organization. This is based on the assumption that experiences in working with others 13 on.projects might carry over to the interviewing situation. .A hypothe- sis to this effect follows: Interviewers who have not had experiences as members of an organization report_problems of communication and gpposition less frequently than do those who have had this experience. Similarly, information about the interviewer's experience in com- munity activities was requested. It is considered that the interviewers who have had previous experience in participating in community activities will have less difficulty interviewing because of these experiences. The possibility of having greater "know how" in selling the idea of such a project, and the Opportunity to learn the skills of how'to overcome re- sistance to community projects, might be increased by participation in these local activities. If these skills are so learned, they might be transferred to the interview situation, and be reflected in the inter- viewer's ability to interview with few problems of communication.and opposition. There is no evidence upon which to base this; it is assumed from Observations of the author's. There is another consideration which supports a hypothesis suggest- ing a relationship between participation in community activities and interviewer problems. This is the consideration that certain of the interviewers who have participated in community activities will probably, through this experience, be involved in a wider network of obligations than those who have not participated in community activities. Whether the informant knows the interviewer or not, he may know of him or of his work in the community and feel a certain obligation to c00perate with 1h him because of his community mindedness, that is, unless the informant has a negative Opinion of the interviewer. Knowing this about the interviewer makes the relationship between the interviewer and the reapondent different than between an interviewer and an informant who know nothing about each other. If the reSpondent feels obligated to c00perate with an interviewer who is known to participate in community activities, there is a possibility that fewer problems would arise. For this reason and the above, the following hypothesis is suggested: Interviewers who have had previous experience in participptipg ,‘in.community activities report_problems of communication and ‘\/ ppposition less frequently than do those who have not had such experience. Five hypotheses have been listed above which are concerned with the previous experiences of the interviewers. They will be tested in Part B of Chapter V. Involvement and.Participation of the Interviewers in the Survey There are several experiences which the interviewers might have had either in becoming involved in the survey or during their training and interviewing which are considered to have a relationship to the types of interviewer problems reported. These are l) the method of township organization for involving the interviewers in the survey} 2) the inter- viewer's position in the survey organization; 3) his training for the interviewing job; h) the number of questionnaires which he completed; 5) the extent to which he found that the respondents had heard about the survey before he arrived for the interview; These variables form the 15 basis for hypotheses as will be outlined in the following: 1) The local township organizations involved the interviewers in the survey along three general lines, as indicated on Page 3. These were lines of friendship, formal organization, or residence in a school district. ‘Whether the method of their involvement has any relationship to the types of interviewer problems reported can at best be only supposed. There is a possibility that there are very different feelings of obligation which are called forth when these lines of involvement are utilized. One's obligation to his friend or relative is a very personal thing, depending upon the degree of the relationship. One's obligation to his organization is possibly less personal and depends on the im- portance of the organization to him. Obligation to the person to his community is the least personal, and his motivation to participate in community activities probably varies with the degree of identification with the community. There prObably is a great amount of variation among the interviewers in the degree to which they feel obligation to the source of their involvement. There may also be a greater “within group" variation than a "between group” variation, particularly because not all of the interviewers were involved along the lines identified with their township or village organization for the survey. However, the author would like to determine if there is a significant “between group" variation when these categories are compared to the types of inter- viewer problems. The following hypothesis therefore is prOposed: There is a difference among distributions of interviewers by the method of township_opg§nization in the extent to which they report problems of communication and apposition. l6 2) Some of the interviewers were also local organizers for the survey. By their contact with county organizers or other people associ- ated with the planning and organizing of the survey, and by their exe periences in involving others to work on the survey, they had the opportunity to learn the purposes of the project, to relate themselves to more than the local aspects of the survey, to practice telling others about the survey when they asked people to volunteer for the interview- ing job. There is a possibility that this added knowledge and experience would facilitate his interviewing, eSpecially in overcoming problems and in explaining the idea of the survey to the informant. If this is true, it can be hypothesized that the organizers encountered interviewer prob- lems less frequently: Interviewers who_participated in organizing_the survey localhy reported_pr0blams of gpposition and communication less fre- gpently than did the interviewers who did not have this experi- ence. 3) Interviewer training is almost always stressed in the literature concerning survey methods, and training meetings are the most frequently mentioned type of training. Hadley Cantril published a study in which he attempted to compare the quality of the findings of untrained inter- viewers with those of trained interviewers for public Opinion polls. The results of his study do not support the emphasis other writers have put on the training meeting experience, for he found that there were no differences between trained and untrained groups in the quality of their l7 interviewing which could be attributed to training.9 This is a sig- nificant finding, for it permits one to question why agencies go to the expense of interviewer training if the quality of their work is not related to their training. There is one possible explanation which.can be tested in this study; namely, that interviewer training is related to the types of interviewer problems which the participants in the survey report. There is one reference in the literature to this pos- sibility. It is a finding from a study carried out at Ohio State Uni- versity; A group of student interviewers were asked to check off the types of problems which they encountered during the survey. It was re- ported that the group of students who had been in the class which received the most careful and complete training encountered fewer prob- 10 This finding lends support to the following hypothesis concern- lems. ing interviewer training: ,' Interviewers who had been trained report problems of communi- '\\jfl cation and opposition less frequently than do those who have not been trained. h).Another variable which might be related to whether or not the interviewer reports that he had interviewer problems is the number of interviews which he completed. There is a possibility that most of the interviewers had difficulty with their first few interviews, but as the 9. Hadley Cantril. Gua Public Opinion, (Princeton: Princeton University Press, 19 , p. 96. 10. Franklin H. Knower. ”An.Inventory of Public Opinion Pollers' Inter- viewing Problems' International Journal of Opinion and Attitude Research, S (1951;, p. 223. 18 job became familiar, they forgot that they had ever had any difficulty. The other possibility is that the more interviews which an interviewer completed, the greater was his chance of calling upon someone who was apposed to the whole survey idea and thus created a problem for the interviewer. Both of these possibilities can be tested. By noting the direction of the association11 (assuming that an association is found), one could obtain an indication of which of the possibilities is the more likely one. The hypothesis is stated in such a way as to specify no particular direction: There is no difference amongpdistributions of interviewers by the number of interviews completed in the extent to which they repprtpproblems of communication and opposition. 5) The final variable in this group is concerned with the extent to which the informants had heard about the survey before they were approached for an interview; Favorable publicity for the survey was pro- vided by the local and county presses, and most of the peOple who had heard about the survey before they were interviewed reported that they had read it in the newspaper. In Lenawee County, it appears that the newspapers are particularly in tune with the sentiments and ideologies of the citizens so that it is probable that the majority of the pe0ple who had heard about the survey had acquired a favorable attitude from 11. Direction of association is a statistical term which.means that in a contingency table, one can indicate with a plus or minus sign whether a given observation is larger or smaller than would theo- retically have been expected. If the observed frequency is larger than the theoretical frequency, the direction of the association is said to be positive. 19 the presses, and therefore should be less likely to be Opposed to being interviewed. Likewise, having read about the survey in the paper, they should be more familiar with its purposes and in general with the types of questions contained in the survey form. It is these assump- tions which led to the formulation of the following hypothesis: The volunteers who interviewed respondents who had heard about the survqy'before the interview report problems of communi- cation and Opposition less frequently than those who did not. The five hypotheses which have been outlined above will be tested in Part C of Chapter V. C. Clarification of the Terms Used in.This Study The survey form used in the Lenawee County self-survey was not what is usually considered as a questionnaire, but is rather an interview schedule. A gpestionnaire is constructed in such a way that the informant is capable of filling in the answers himself. A schedule, on the other hand, is designed in such a way that an interviewer is needed to admin- ister it. Because it will be necessary to refer to two different schedules in this study, one will be referred to as a questionnaire and the other as a schedule. The Health and Defense Survey form used by volunteer, non-professional interviewers to interview residents of Lenawee County will be called a questionnaire. The survey form con- structed for the use of staff members of the Social Research Service of Michigan State College in finding out about the training and involve- ment of the volunteer interviewers in the Lenawee County survey will be referred to as a schedule. 20 A similar problem in nomenclature exists between the volunteer interviewers in.Lenawee County and the paid, professional interviewers from Michigan State College who interviewed the volunteers. The volun- teer, non-professional interviewers will be referred to in this study as the interviewers, the participants, or the volunteers. The paid, professional interviewers will be referred to as part of the research team. CHAPTER II REVIEW OF LITERATURE This thesis problem brings together three areas of the literature concerned with social surveys. First, it considers interviewer problems. Secondly, it is concerned with the employment qualifications of research agencies in regard to the age, sex, education, occupational experience, etc. required for an interviewing job, and the training given new inter- viewers. Finally, it is concerned with the interviewers who participate in self-surveys. Each of these areas will be treated separately first, and then summarized in terms of their impertance to this thesis. ‘Where a particular reference has formed the basis for one of the hypotheses stated in Chapter I, this fact will be indicated. A. Interviewer Problems Since interviewer problems are the main emphasis of this thesis, they should receive the most attention in the Review of Literature. However, little attention has been paid to interviewer problems in the literature concerned either with professional or self-surveys. Survey agencies, such as the National Opinion Research Center, have concentrated a great amount of attention on interviewipggprcblems, that is, problems of the agencies or organizations engaging interviewers, but little atten- tion to intervieweg_problems. The concern with interviewing problems 22 may be the result of their being a great expense to agencies; the lack of attention to interviewer problems before this time may be due to the fact that it has only recently'been.realized that interviewer prob— lems are also an expense to the agency in.that they cause errors in the findings of the survey. Franklin Knower attempts to explain this in his article in the International Journal of Opinion and Attitude Research. He says that interviewing which is not skillfully done is usually not accurate and complete, and is therefore an expense to the agency or voluntary organization. He defines unskillful interviewing as interviewb ing in which significant problems arise that the interviewer is not prepared to solve.1 In the same article, a study is described which is concerned with these questions: ,Are there common problems which rela- tively inexperienced interviewers encounter? Can skill in interviewing be measured by the frequency with which problems are reported? Can interviewing skill be deve10ped by training? The committee which cone ducted the research.designed to answer these questions found that certain problems were actually encountered very frequently by inexperienced interviewers. These were: difficulty in.making the interviewee under- stand the questions; difficulty in keeping the interviewee from wandering from the point; difficulty in gaining the consent of the reSpondent to be interviewed; difficulty in getting the interviewee to make the mental effort required to answer the questions thoughtfully, etc. They found that skill in interviewing is apparently a general measureable achievement, l. Franklin.H. Knower. “An.Inventory of Public 0pinion.Pollers' Inter- viewing Problems', International Journal of Qpinion and Attitude Research, 5(1951), p. 221. 23 and that interviewers vary in the difficulties they encounter as their maturity, training, and skill vary. In the conclusions, Kncwer points out that successful interviewing demands Specific training to meet specific problems that arise in different types of projects.2 It is interesting that two of the types of problems described by Knower as being the most frequently encountered are the problems used \ as variables in this study, problems Of Opposition and problems of x / communication“ .As in this thesis, the prOblems are reported problems of the interviewers. Likewise, the finding that training is related in \ some way to interviewing skill, measured_in terms of the frequency with x' I ‘\, which the interviewer reports problems, is added support to the hypothe- / \ sis suggesting a relationship between training and interviewer problems mentioned on Page 17 above. The two types of problems discussed in this study each.are referred to in other writings in the literature. Leo Cherne refers to the pos- sibility of interviewer encountering problems of Opposition when he writes, The interviewer must be trained in advance to expect refusals to answer. He must know that his temper will be aroused, be- cause only if he knows this beforehand will he be able to keep it in restraint, and he must keep it in restraint, if he is to overcome attempts by the re3pondent to Oppose his efforts to complete the interview satisfactorily; He must be cautioned, and he must be taught the verbal techniques by which to avoid argument and by which to give the appearance of agreement with- out feeding the passions Of prejudice. A practice interview in the training room with Opportunities to overcome Opposition 2. Ibid., p. 228. 2h may provide the interviewer with greater knowledge of the role he is going to play.3 As can be seen from the statement, Cherne also recommends a training session and a practice interview as methods for learning to overcome Opposition. The hypothesis to this effect is stated on Page 17. Luther Fry reports a study in which failure Ofthe interviewers to communicate with the reapondents resulted in highly distorted results. Textile workers in New Bedford, Connecticutt, were on strike in 1928. At that time, they were interviewed to find out if they favored arbitra- tion. Eighty-five per cent of them replied that they did not. On further investigation, it was fOund that they thought that the word "arbitration” meant "surrender" or "resignation". A resurvey using a different word showed much different results.Ll . Failures to communicate frequently are not recognized either by the agency or by the interviewer unless the respondent asks a direct question about the meaning of a schedule item or a particular word. This is probably one of the reasons that such difficulty is not fre- quently referred to in the literature. B. Qualifications for Interviewers The qualifications prescribed by various agencies for employment as an interviewer vary considerably. Thar; is little agreement on age, 3. Leo Cherne . To Fulfill 0x151 Promises-~An Examination of Communigy Audits? National Citizens Council on Civil Rights, New York, p. 13. h. Luther Fry, The Techniques of Social Invesggation, (New York: Harper and Brothers, 1931;) , p. 35. 25 sex, education, occupational experience, etc. There is some evidence, however, that the nature of the Specific interviewing job, rather than any generally agreed upon requirements for the interviewers, prescribes the qualifications. Some of the statements made by the specialists in social research and opinion pollsters are included below to demonstrate the variation in the prescribed qualifications and to show wherever possible variation which is based on the type of the survey. Social Characteristics-Age The range of age requirements for interviewing is from elementary school children to men and women as old as 60 years. The great age range permitted is probably a function of the nature of the surveys and the age group for which they were designed. For example, elementary school children in Petersburg, West Virginia, with the assistance of their teachers, carried out a self-survey of health and sanitation prob- lems. Their findings were complete enough to be accepted by civic groups after informal verification of the data.5 Similar self-surveys have been conducted easily by youth groups concerning the employment possibili- ties, the housing facilities, the recreation prOgram, etc. of their home town.6 However, the majority of the surveys, whether they were done by professional or volunteers, are planned and conducted by adults . S. Effie Bathhurst. Petersburg Builds a Health Program, Federal Securi- ity Agency, Bull. 9, 19149, United States Government Printing Office, Washington, D. C. 6. Ann G. Wolfe. Teen-53ers Look at Their Town, Division of Youth Services of the American Jewish Committee, New York. 26 Archibald Bennett points out that public Opinion agencies employ prima- ily interviewers between the ages of thirty and fortybnine years. If they are less than thirty, they are employed because they are college graduates and therefore are considered to be better qualified.7 Inter- viewers who are more than fifty years old are considered to have more difficulty in "getting around“, are more affected by bad weather, are slower, etc. Some writers suggest that older peOple have such set ideas that they cannot help but bias the findings. BOgardus suggests this when he writes: Older mature persons, who 'volunteer' do not as a rule make satisfactory research workers. They are likely to have some preconceived notions on 'how the thing ought to be done' which cuts down their usefulness in scientific researcht..They have difficulty in.making the scientific approach and in getting back to fundamental ecolOgical bases. Their interest in 'facts' and the formal side of the data hinders them greatly in getting behind the data and examining the processes. Their lack of understanding of the nature of attitudes and Op ons is a serious handicap in explaining the survey to others. Summarizing the above, it would seem that only the upper age groups are considered to have questionable success in interviewing. The hypothe- sis based on these writings is stated on Page 9. Social Characteristics--3ex Interviewing is considered by many authors to be primarily a woman's job. This is partly due to the fact that paid-interviewing is frequently a part-time, temporary rather than a long-range, permanent type of 7. Archibald Bennett. Report on.Researching Researchers, (New York: Bennett.Associates, l9h87:7p. 23. 8. Emory S. Bogardus. The New Social Research, (Los.Angeles: Jesse Ray Miller 00., 19267, p. 56.— 27 employment. women likewise are called upon more frequently to do voluntary interviewing, because they can free themselves from their housework for longer periods of time than men can from their businesses. The fact that women can participate in interviewing more conveniently than can men is one side of the question, but what is more interesting is that women are preferred by many agencies as interviewers for reasons other than their availability and their acceptance of lower wages. Holoneqyx‘points out that women are less inclined to cheat than men, and that women prove to be better for part-time interviewing, especially on readership studies. He also states that men are better interviewers for crew jobs, and that the ideal agency staff should include approxi- mately twenty percent men for this reason as well as for supervisory jobs.9 Edith Witt suggests other reasons why agencies prefer women inter- viewers. These were referred to already in Chapter I as being: easier access to homes, stores, business, etc. , and availability for part time work with, relatively low pay}0 The conditions suggested in this refer- ence may account for some of the refusals and Opposition reported by the male interviewers . Social Characteristics-:Education The policy of various survey agencies regarding the amount of education needed to interview varies somewhat with the difficulty of the 9. John Maloney. "Interviewing Problems" , Central City Conference on Public Opinion Research, National Opinion Research Center, University of Denver, Denver, Colorado, l9h6, p. 22. lo. Witt, 92. 933., p. 1:39. 28 interview schedule and the type of information sought. The National Opinion.Research.Center maintains two staffs of interviewers. One group is composed almost entirely of college trained persons. Their assignments are usually of the depth interview varietyb-they use schedules which are primarily "Open-ended", aimed.at getting at peOple's opinions of national policies.. The other group represents a cross section of the populations based on education, occupation, religious affiliation, etc. They are employed primarily for the less complicated public Opinion polls, such as pre-ele¢tion.polls, which aim at getting a wide coverage of the pOpulation. George Lundberg represents a similar viewpoint that interviewers should have the amount of education necessary to do the job. He states: If the inquiry is of a specialized nature so as to require the services of Specially educated workers, these should be selected as Specialized talent is for any work. If the field work can be performed by persons without education above high school, pro- vided they are carefully instructed, and the necessities of the investigation require the utilization of such persons, the most important criterion of their desireability is that theyishould be able and willing to follow instructions explicitly. Laszlo Radvanyi reports the opposite. He states that regardless of the type of the survey, he prefers college students as interviewers, as long‘as they possess a keen scientific interest in research. He re- ports that interviewers with less education turn in a low calibre of work, and that interviewers who are college graduates expect too much salary and are seldom interested in the interviewing job.12 11. George A. Lundberg. Social Research, (New'York: Longmans, Green, and.Co., 19h2), p. 353. 12. Laszlo Radvanyi. IInterviewing Problems", Central City;Conference on.Public Opinion Research, National Opinion Research Center, University of Denver, Denver, Colorado, 19h6, p. 26. 29 Kathleen Larkin has had considerable experience in selecting volunteers to do an interviewing job. She recommends that 'peOple with a high school education" make better interviewers. {About an individual volunteer who has less than a high school education, she says, ”There is no sure way to size up the volunteer who has had only an eighth grade education. Interviewing requires training--the learning of a new skill. Can he come through?"13 Likewise, she finds that most of the volunteer interviewer jobs for which she selects peOple are simple enough that the college trained individuals feel that it is beneath their skills, so that for her the high school educated individual.makes the best inter- viewer.1h Summarizing the above, it appears that although a college education is preferred by most public Opinion agencies, interviewers with less education are hired, and in the case of at least one voluntary agency, interviewers with only a high school education are preferred. There are no references which attempt to show if there is any relationship between the amount of education experienced and the number of interviewer prob- lems encountered. Education is a factor which has been shown to be related to the quality of an individual's interviewing but not to his interviewing problems. 13. K. O. Larkin. For Volunteers Who Interview, Volunteer Bureau Welfare Council ofVMetropolitan.Chicago, Chicago, Illinois, l9h9, p. 12. 1h. Ibid., p. 1h. 30 C. Previous Experiences of the Interviewers Occppation There is no reference made in the literature which would indicate that interviewers are selected on the basis of their previous occupa- tional experience. However, there is an idea which seems to occur in many of the writings that peOple who are interested in certain aSpects of community life, such as teachers, social workers, and college stus dents prove to be earnest and capable field workers. Likewise, Lundberg suggests that those people who are interested in the social sciences generally are more successful as interviewers than those who are not,15 while Radvanyi indicates that it is the scientific interest of the individual which is the discriminating factor.16 Lester Guest attempted to test the hypothesis that interviewer excellence is related to inter- est, as reflected on the Strong Vocational Interest Test. He found that individuals who had a vocational inclination toward the ministry, social science teaching, and writing received the highest ratings in excellence of interviewing.17 The above might also be true for interviewer problems, and certainly the literature above would support a hypothesis to this effect. There is no data in the Lenawee Study which could be used in testing such a hypothesis , however . 15. Lundberg, SE~.EEE-: p. 353. 16. Radvanyi, 22. 933., p. 22. 17. Lester Guest. "A Study of Interviewer Competence", International Journal of Opinion and.Attitude Research, 1(l9h7), p. 279. 31 Previous Interviewing Expgrience The literature Offers no substantial evidence concerning the effect of length of experience on interviewer performance. While it would seem logical that increased experience would improve an inter- viewer's work, many research workers claim that the more experience the interviewer has, the more he has learned about short cuts and how to cheat without detection, and he may have developed idiosyncracies of interviewing as well .18 Others, for example, Archibald Bennett, con- sider that experience is tantamount to interviewing skill .19 The policy of the research agencies, National Opinion Research Center, GaIIOp Polls, and Roper Polls in particular, supports the first view. They do not hire interviewers, unless there is a shortage of new applicants, who have had a great amount of prior experience. In general, they prefer to hire inexperienced applicants, because they consider that they will have a “fresh approach to the work, will be highly motivated, and can easily be trained with the agency's techniques."20 There is only one study which has been carried out to test the effect of previous interviewing experience on the quality of the inter- viewer's work. This study was carried out by Paul Sheatsley. He found that there was no significant difference between interviewers who had 18. Mosteler, Hyman, McCarthy, 3311., The ____Pre-E1ection Polls of 191:8, Social Science Research Council, New York, 19149, p. 1146. 19. Bennett, 22, g;3., p. 3b. 20. Paul Sheatsley. "Interviewer Characteristics and Performance, Part III', International Journal of Opinion and Attitude Research, 5(1951), p.199. 32 had from six months to five years experience and those who had had no previous experience in the quality of their interviewing. He did, however, find that those interviewers who had been on the job for more than five years were rated "below average" significantly more times than those who had had less experience.21 It is the writer's Opinion that in the case of interviewer problems, the effect of experience would not be a negative one as it appears to be in the case of interviewer excellence, but rather that once he had learned to overcome opposition, to communicate the meaning of the ques- tions, he would not forget these skills, while he might tend to become careless in his probing and not get as complete and accurate answers. Therefore, the hypothesis suggesting a positive relationship between experience as an interviewer and a minimum of interviewer problems is not based on the literature cited above. (See Page 11) D. Interviewer Training The desirability of interviewer training is stressed everywhere in the literature, but the types of training specified varies to some de- gree. Some of the reasons given for the value placed on interviewer training are that well trained interviewers are better able to probe objectively, that training helps familiarize the interviewer with the schedule, that training brings‘about confidence, that training can make the interviewer see his relationship to the whole project, etc. These 21. Paul Sheatsley. "Interviewer Characteristics and Performance, Part II", International Journal of Opinion and Attitude Research, h(1950), p. 20h. 33 are conclusions which have been derived from experience, it would ap- pear, rather than from research, Few studies have attempted to test ‘whether interviewer training actually has the stated effect upon the interviewers. The two studies which have been attempted in this area were carried out by Cantril and Knower. In.Cantril's study, two groups were compared as to quality of interviewing. One group is called the ”untrained" group; the other is referred to as the "trained“ group. Actually, the first group was sent by mail a copy of the instructions for carrying out a particular Opinion survey; the second group was trained by National Opinion Research Council supervisors in a regular training meeting with.practice interviews. ‘Hhen these two groups were compared as to quality of their interviewing statistically, no differ- ence was observed which could not have occurred by chance. It should be pointed out, however, that the results of this study did not change the policy of NORC which makes a point of personally training all of its employees.22 ' The study which is of even greater relevance to this thesis has already mentioned in the section of Interviewer Problems. In this, Knower reported that the students who received the most training for an Ohio State University study also encountered the fewest interviewer problems. He suggests that training pragrams should be based on a consideration of the problems which interviewers encounter--that one's 22. Hadley Cantril. Guaging Public Opinion, (Princeton: Princeton University Press, l9hh), p. 96. 3h 23 failure to overcome problems affects greatly the results of a survey. E. Community Self-Surveys The reader is probably aware of the fact that most of the litera- ture cited above concerns paid interviewers for professional surveys. This is primarily due to the fact that, although many reports have been prepared concerning the findings of self-surveys, few authors have con- sidered the organization of a self-survey, the volunteer interviewer, his training, etc. It is the writer's Opinion, however, that interviewb ing for a self-survey requires almost as much skill as is required for most professional social surveys. It is for this reason that the litera- ture concerning professional surveys, as well as selfbsurveys, has been used to form the bases for the hypotheses to be tested with the data collected about interviewers fer a self—survey; It is considered necessary, however, to point out what is meant by a self-survey and wherein it differs from other social surveys before proceeding with this study. The following is a review Of the literature concerning the community self-survey. Definition of the Term "Self-Survey” The term "self-survey" refers to surveys "in which citizens of a community gather facts about some social problems, and which.aim to bring about subsequent action to alleviate that problem.2h’ It is not 23. Knower, 92. 533., p. 223. 2h, Margot Haas'Wbrmser and.Claire Selltiz. "Community Self-Surveys," Jahoda, Deutsch, and Cook, Research Methods in Social Relations, (New York: Dryden Press, 19517, Vol. II, p. 612. 35 known to the author just when the first self-surveys were conducted, for many surveys of the self-survey nature were carried out before the development of the term itself. The Douglas County Rural Youth Survey of 193).;25 and the Ross County Rural Youth Survey of 1911.126 were both planned and carried out by local planning committees. The emphasis of these, like the current self-surveys of health, discrimination, recrea- tion, etc. was not only upon the collection Of facts, but also on the action programs which might result. This is the important fact about self-surveys up and above their local character: they are action oriented. As stated by Homer and Selltiz , the basic assumptions of the self-survey W method outlines below will show this to be true: 1. Individuals who participate in an undertaking tend to become ego-involved in it. In a fact-finding project, such ego- involvement is likely to take the form of a sense of personal responsibility to do something about the situations which one has helped to discover, if those situations do not conform to general standards of equitable social arrangements . 2 . Group membership reinforces individual commitment and pro- vides support for individual behavior, with the result that people who may lack courage or skill to work by themselves toward changes which they believe to be socially desirable are able to undertake constructive action as members of a group. Through its use of working committees and other Operating groups, the self-survey brings to bear these group membership effects. 3. In any self-survey focused on problems over which there are apparent conflicts in interest between sub-groups in the com- munity, the experience of members of these various sub-groups 25. Douglas County Rural Youth Survey, Rural SociolOgy Department, Agricultural Extension Service, University of Wisconsin, 1935. 26. Ross County Rural Youth Survey, Ross County Rural Planning Committee , 19141. 36 in working together on a study of the common problems may itself help change attitudes which have contributed to the problem.27 Whether these assumption have proved themselves in experience has not been shown experimentally. Each report of a self-survey gives primarily the findings concerning the problem; a few give accmmts of action allegedly resulting from the survey. Let's Look at Ourselves, for instance, states that after the Montclair Audit, Negro physicians were appointed to the local hospital staffs for the first time}:8 Have You Heard What's Cooking? reports that ”Resolutions pledging equal treatment of all patrons were received from unions representing more than 70,000 employees and from management associations comprising more than 1500 restaurants.”29 Up to the present time, the author has not been able to locate any follow-up studies which actually have shown that changes came about in the comunity as a result Of a self-survey. However, Richards Davids found that shortly after the completion of a self-survey of health in Clinton County, Ohio, a tremendous interest in health and health activities arose among the peOple. Two well-baby clinics were started; compulsory imrmmization as a school entrance re- quirement was adopted; a prOgram to X-ray all high school children for tuberculosis has been instigated; classes for expectant mothers have 27. worn-er and Selltiz, 22. 923., p. 612. 28. Let's Look at Ourselves, National Citizens Council on Civil Rights, NewYork, p. 11. 29. Have You Heard What's Cooking? , American Jewish Committee , New York, p. 7. - 37 begun. All these actions have occurred since the survey.30 A further assumption of the self-survey method is that peOple in the community who have not participated will be willing to accept the findings of a survey carried out by peOple in their own community more frequently than they will the findings of a survey carried out by out- siders who have no legitimate interest in their community.31 Whether this assumption proves correct depends a great deal on the effectiveness of the interviewer as a public relations agent .32 His contacts with the local householder can sell the survey, the findings, and the resulting action program, or they can have a negative effect. Studies regarding the public image of a self-survey have not been located by the author. Recent Self-Surveys Community self-survey with the aims described above appear to have been a major concern of the Race Relations Department of Fisk University since the Second World War. The idea of the self-survey method was ap- plied to inter-group relations by Dr. Charles S. Johnson, President of Fisk University, and the staff of the Race Relations Department in 19146. This group Of social scientists consulted with local groups in 30. Richard Davids. ”Your Answer to Health" , Farm Journal, November, 1951, p. 31-2. 31. Margot Haas Wormser and Stuart Cook, "The Use of the Comunity Self- Survey in Combatting Discrimination" , paper read at the Fifty-fifth Annual Meeting of the American Psychological Association, September, 1910, Detroit, Michigan, p. 2. 32. Cherne, Q. _c_i_t_., p. 11. 38 San Francisco, California,33 Pittsburgh, Pennsylvania,3h Kalamazoo, Michigan,35 and,Minneapolis,lMinnesota,36 and assisted them in carrying out self4surveys of discrimination. Following close behind these came the Northtown Self-Survey of discrimination for which Margot Haas ‘HOrmser of the American Jewish Congress served as consultant,37 and likewise, the audit conducted at Montclair, New Jersey38 which was Ob- served hy the National Citizens Council on Civil Rights, came shortly after. Recent interest in health self-surveys began in 19h? when 317 volunteer interviewers surveyed h,789 Columbiana County, Ohio, families concerning their health problems}? Clinton County, Ohio, citizens followed close behind in l9h8-9 and carried outra.self%survey'of health needs which reached 85% of the population of the county.ho .A self-survey 33. The Negro'WarJWOrker in SanFranciscOJ A Local Survgy, Race Rela- tions Department, Fisk University, Nashville, Tennessee. 3h. Pittsburgh Stugy, Race Relations Department, Fisk University, Nashvil 9, Tennessee. 35. Nggros in Kalamazoo A Local iself-Survey, Race Relations Department, Fisk University, Nashville, Tennessee. 36. Minneapglis Community Sglf-Survgy, Race Relations Department, Fisk University, Nashville, Tennessee. 37. Margot Haas WOrmser, Northtown Survey;of Human Relations, Commission on.Community Interrelations, timerican Jewish Congress, l9h7. 38. The Montclair.Audit, Montclair Intergroup Council, Montclair, New Jersey, 19H9: 39. You.and Your Neighbor, Agricultural Extension Service, Ohio State University, bull. 307, April, l9h9. hO. Clinton,County:Health Survey, Clinton.County‘Hea1th.Council, Wilmington, Ohio, 19§o. 39 can be done by young peOple as well as by adults. Elementary school children in Petersburg, west Virginia, located the health problems in their town.141 Likewise,.Ann Wolfe described a self-survey of health, recreation, sewage diaposal, etc. carried out entirely by teen-agers.242 Although these surveys are action oriented and the aim primarily is for community improvement rather than.simply fact-finding, it should be pointed out that to the extent that self-surveys are methods of fact- finding just as is true of any type of social survey, the principles and techniques which have meaning for professional survey agencies are likewise applicable in interviewing for a self-survey. The schedules used by interviewers in these communities are nearly as complicated as most professional interview forms. The problems encountered by the volunteer interviewers are reportedly the same as those encountered by professional interviewers. The difference between self-surveys and professional surveys as far as interviewing is concerned is that the interviewers in a self-survey are volunteers and are usually inexperienced. The manuals which are beginning to appear on the market concerning the methods for carrying out a self-survey in one's local community suggest the same training for volunteer interviewers as do the professional hand- books for agency employees}43 bl. Bathhurst, 22. cit. 142. Wolfe, 22. 312. A3. An.example is: ‘Margot Haas WOrmser and Claire Selltiz, How to Conduct a Community Self-Survey of Civil Rights, (New York: Association.Press, 1952166 CHAPTER III METHODOLOGY Chapter III is concerned with the method used in the collection and analysis of the data which form the basis for this thesis. Attention will be given to the construction of the interview schedule, the testing of the schedule, the population interviewed, the interviewers, and the statistical techniques employed in the analysis of the findings. A. Interview Schedule Data for testing the hypotheses outlined in Chapter I were collected by means of an interview schedule administered to 313 of the volunteer interviewers who participated in the Lenawee County Health and Defense Survey. An.attempt was made to interview all the participants in the survey to obtain infOrmation concerning their involvement in the survey, their attitudes toward the survey, their previous interviewing or sales k»... Hiya-1,...“ experience, their opinions as to what might result from the survey, their w interviewing problems, their training for the interviewing job, their .- ' ...... m. age, sex, education, occupational experience, etc. A formal interview sshedule was constructed by members of the staff of the Health Informa- tion.Foundation financed project carried out by the Social Research Service of the Department of SociolOgy and.Anthr0p010gy at Michigan State College. (See page 3). Al The particular questions which were utilized in the develOpment of this study are listed below; a COpy of the schedule can be found in Appendix B. The numbers beside each of the questions are those which can be located on the interview schedule. 5. Address If in Open country: If in town or village: 6. Sex of respondent Male Female 9. About how many interviews did you take? Towns hip School District name/and/or no. Town Street 10. (a) Had you ever interviewed before? i.e. , in such things as: (b) Have you ever done any door to door selling? (1) Yes (2) No __ U. S. Census School Census Church Census Social Welfare Worker Other No (c) If yes, about how. many house calls did you make in this selling? (1) Under 20 (2) 21-100 (3) Over 100 11. Can you tell me in general how you were given instructions on how to do the interviewing job? 12. Do you feel that you received sufficient training for the interview- ing job you did on this study? (Give time for thinking beyond a mere yes or no, eSpecially if there is hesitancy) . A2 13. Do you feel that you received sufficient assistance in doing the interviewing job? Did you have problems or questions, for in- stance, and did someone help you with them? 114. (Hand interviewer (informant) a COpy of the self-survey question- naire) On which questions did you find it most difficult to get answers? (1) Why did peOple find these questions difficult? (2)’Were there any questions which you did not understand? (3).Any which you didn't understand why it was asked? (A) Probe for: a. problems of memory b. problems of “blockage" c. problems of evasion d. problems of communicating the meaning 1?. Of the families you approached, how many refused to be interviewed? What was the common reasons for refusing? 18. About how may of the peOple you interviewed had heard about the ‘ survey before you arrived for the interview? b3. About how long have;you lived in this county? Under 1 year One to five years , Five to ten years , Over ten years— h6.‘What was the last grade or year you completed there (last school)? 1-14 No. of years in college, 537 if not a graduate 8 Professional Scmol (Specify) Some high school Kind and Length of Training High school graduate Other 147. Would you mind telling me how Old you are ‘ (Age range--ten year gapST 148. Are you married? Married Divorced Widowed Never married 149. (If reapondent is male, ask:) What is your major occupation or profession? About how long have you been in this occupation? What other occupations have you had any considerable length of time? 50. (If reSpondent is female, ask:) What is your husband's major occu- pation or profession? About how long has he been in this occupation? What other occupations has he had for any considerable length of time? A3 51. (If reSpondent is male, ask:) Does your wife have an occupation or profession outside the home? If yes, What? How long? 52. (If respondent is female, ask:) Do you have an occupation or pro- fession other than homemaking? If yes, What? How long? If no, have you ever had an occupation or profession outside the home? If yes, What? How long? 53. Would you mind telling me what religion you consider yourself? 55. Besides the church, what organizations are you most active in? 56. Have you been in the past very active in other community projects such as the health survey? (Probe for details on civic activity in the past, also Specify kinds of activity) It should be pointed out that the schedule had already been con- structed, tested, and much of the interviewing had been completed before the author designed this thesis project. Therefore, there are several questions which might have been added had a problem been selected before the schedule was constructed. B. Testing of the Interview Schedule One township completed its self-survey in August, 1951, while the other townships and local communities did not begin theirs until September or later. Therefore, the interviewers in that township afforded the re- search staff a small sample upon which to test the interview schedule. Approximately twenty of the volunteer interviewers were themselves inter- viewed in a test of the schedule. A few changes were made after the findings were analyzod, and a few questions were added to the interview form . C. Population There were an estimated 382 peOple who participated in the Lenawee County Health and Defense Survey in the rural areas.1 Of these, 326, or 85%, were interviewed. Those not interviewed were finally omitted because of technical problems of weather, bad roads, not home even after call back, etc. Of the 326, there were 313 who actually did some of the interviewing in Lenawee County. The other thirteen partici- pated in the survey as organizers. Since this study is concerned only with the problems of the volunteer participants who did some of the interviewing, only the schedules of these 313 have been analyzed. The coverage per township varied from 57% to 100%; this is considered ade- quate for this research. D . The Interviewers The peOple who interviewed the volunteer participants in Lenawee County were college trained, paid employees of the Social Research Service of the Department of Sociology and AnthrOpology at Michigan State College. Some were graduate students in Sociology; the remainder had received their college training in other fields. The author of this study was one of the paid interviewers of the project. The group was trained and supervised by staff members Of the Social Research Service. 1. Approximately 200 interviewers took part in the Tecumseh and,Adrian surveys, but information concerning their interviewing experiences was not available at the time this study was undertaken. 115 E. Developing the Code and Analyzing the Data The code was designed by the research staff after much of the interviewing had been completed. The codes for the Openpended questions were based on these returns. ‘When the coding was completed, the data were punched on IBM cards. The author functioned as a member of the research staff in the code de- velOpment and had access to a set of the project IBM cards. F. Chi-Square Method of.Ana1ysis Once the IBM cards had been run.and the data obtained from the cross- runs, statistical tests were applied. Specifically, the Chi-Square method of statistical analysis was used to test the various hypotheses. The Chi- Square method is a test for contingency; That is, it is a statistical test used to find out if two variables are related in other than a chance manner. In the case of the Chi-Square method, both variables are 222? guantitative. This means that they are capable Of description in terms other than numerical. Examples of nonpquantitative variables are: sex, marital status, occupation, etc. The data for this study are particularly suited for such analysis, for they are primarily nonequantitative. This is frequently the case with sociological findings in contrast to data in other fields where devices for more precise measurement have been de- veloped. The Chi-Square method employed in this study is the one described in Margaret Hagood's Statistics for Sociologists.2 The procedures fOr 2. IMargaret Hagood. Statistics for SociolOgists, (New'York: Henry Holt and Co., l9h0), pp. h88-5h0. A6 demonstrating the existence of association, the direction of associ- ation, and the degree of association between two variables are described there. The significance test used to demonstrate the existence of association necessitates the use of a table of Chi-Square (X2) distribus tion developed.by Fisher and Yates,3 Hagood's textbook.h The direction of association can be shown with S and reprinted with permission in a contingency table pictured by Hagood, while to show the degree of association necessitates that one be able to apply the formula for co- 6 efficient of mean-square contingency. C-values can.be corrected so that they can be used to compare the degree of association for tables having different numbers of cells. McCormick has supplied the formula and table for the Operation.7 3. R. A. Fisher and F. Yates. Statistical Tables for Biological, Agricultural ,_and Medical Research, (London: Oliver and Boyd, 1938), Table IV, p. 27. A. Hagood, 32. 2:93., p. 90A. 5. Ibid., p. 52A. 6. Ibid., p. 508. 7. Thomas Carson McCormick. Elementa3y_Socia1 Statistics, (New York: McGraw Hill, 1910), p. 398. CHAPTER IV PROBLEMS REPORTED BY THE INTERVIEWERS This chapter will primarily consider the problems which the volunr teer interviewers stated that they encountered during the Lenawee County Health and Defense Survey. In the previous three chapters, the fact that there were problems of communication and Opposition.encountered by the volunteer participants has been accepted as a given. 'In Chapter IV, however, some of the statements of the participants from which these categories have been derived will be included to demonstrate the reasons for their derivation. ,Also, the questionnaire used by the volunteer interviewers in the Lenawee County Survey will be described, and the items of the questionnaire on which the interviewers reported encounter- ing the most difficulty will be indicated. The questionnaire will be discussed first to form a basis for outlining the problems, since the participant's statements refer quite frequently to Specific Questions. A. The Health and.Defense Questionnaire The questionnaire is divided into two parts. (See Appendix,A, the Lenawee County Health and Defense Survey). The first includes questions about Civil Defense preparedness and Special training which would be useful in caring for atomic bomb victims; the second includes questions about defenses against disease, incidence of disease, and use made by A8 each family of the local medical and health facilities. The two parts were designed in such a way that they can easily. be separated and used individually by different agencies. The respondent '3 name and address were requested on the Civil Defense part, so that those people having special training would be known by the local Civil Defense Organization and could be contacted in case of an atomic bomb attack upon Detroit or Toledo or a similar emergency. This part was intended prinmrily for the files of the Lenawee County Civil Defense Organization. Anonimity was assured on the health part of the questionnaire with the hOpe that the re3pondent's answers would be more complete and accurate because of this. The health part was intended for use by the Lenawee County Health Council and other Sponsoring organizations. A discussion of each of these parts will follow: The Civil Defense Part of the Questionnaire The Civil Defense part of the Lenawee County Health and Defense Survey included questions about each of the four following areas: 1. Military Service-Questions concerning training in de- contaminating gassed areas, training as a Medical Corps- man, a Laboratory Technician, An Army or Navy cook, etc. 2 . Information about Atomic Bomb Attack-Questions asking whether peOple had been informed by pamphlets, movies, etc. about what to do in an atomic bomb attack. 3. Housing Atomic Bomb Attack Evacuee§--Questions concerning a family's ability to house atomic bomb attack evacuees from Detroit and Toledo in case of an emergency. A9 A. Red Cross, Nurse's AideLDietician's Trainllijquuestions about particular types of training which the reapondent or a member of her family might have received previously that would be useful in case of an.emergency. Forty-seven of the 313 volunteer interviewers stated that they encountered problems on the Civil Defense questions. Of these, 39 had difficulty with one of the areas mentioned above, while eight interview- ers encountered problems on the entire Civil Defense part Of the questionnaire. Among the 39 volunteer interviewers who encountered problems on only one area of the questionnaire, 30 of these mentioned specifically the questions having to do with the housing of atomic bomb attack evacuees from Detroit and Toledo. (See Table I) The particular questions referred to read: "In a war emergency, how many extra persons could you house tempOrarily? “Would you prefer ( ) Men. ( ) WOmen ( ) Children. ( ) No Preference?" (See Appendix Most of the interviewers who had difficulty with these questions reported that the difficulty was due to their failure to make the re- spondents understand that they were not obligated to take evacuees into their homes just because they gave an affirmative answer to the above questions, but that the Civil Defense Organization wanted an estimation, not an exact count, of the number of peOple which Lenawee County could be expected to care for in case of an emergency. Likewise, a few of the interviewers reported having difficulty explaining to the reapondent 50 TABLE I DISTRIBUTION OF THE A7 INTERVIEWERS REPORTING PROBLEMS ON THE CIVIL DEFENSE QUESTIONS BY THE AREAS ON WHICH THEY HAD PROBLEMS Areas on Civil Defense Part of Questionnaire Number of Interviewers 1. Military Service (3) xxx* 2. Information about Atomic Bomb Attack:Precautions. (O) 3. Housing of.Atomic Bomb Attack Evacuees from Detroit, Toledo. (30) ‘0‘ ' A. Red Cross, Nurse's,Aide, Dietician's Training, etc. (o) xxxxxx 5. Entire Civil Defense Area of Questionnaire. (8) xxxxxxxx Total A7 a Each.X represents one interviewer just what equipment she should have in her home to take care of an evacuee. In.many cases, the interviewers themselves did not know the answer to this question, and therefore, they could not explain it to the reSpondent. Difficulties on the questions which had to do with particular types Of specialized training, such.as Red Cross, First Aid, Nurse's Aide, home economist, etc., were primarily the result of an admitted lack of knowledge on the part of the interviewer as to how recent the training had to be, or whether high school home economics courses qualified one to be a home economist. Had the interviewer had this information 51 readily available, a decision about the answer to the questions could have been.made easily, and the interviewer would not have identified it as a problem. The majority Of the eight interviewers who reported having dif- ficulty with the entire Civil Defense area Of the questionnaire reported that this difficulty was primarily due to a general Opposition on the part of the respondents to committing themselves to take an evacuee or to participate in helping the community in case of an emergency. There was also some misunderstanding reported as to the purpose of the Civil Defense questions; these peOple indicated that they did not feel that there was a chance of there ever beingian atomic bomb attack, and that therefore such information would.be of little or no value. Others indicated that collecting all that information was a waste_of time, because if there were an atomic bomb attack, it would happen so suddenly that written records would not be consulted during the process of find- ing shelter for the evacuees from Detroit and Toledo. There were other problems which the interviewers reported that they encountered on the Civil Defense questions. For example, three inter- viewers reported having difficulty obtaining answers to the questions about military service, but their reasons were varied, and not signifi- cantly important to be discussed here. None of the interviewers had difficulty with the questions asking about whether the reSpondent had heard about the precautions which should be followed in case of an atomic bomb attack. 52 The Health Part of the Questionnaire The Health part of the questionnaire included questions in eight different areas. These are listed and defined in same detail below: 1. Immunization--Questions about innoculation for small pox, diphtheria, etc. for every member of the family. 2. Past Illnessegr-Questions about the prevalence of mumps, measles, scarlet fever, pneumonia, etc., for each member of the family. 3. Medical Care-Aluestions about the number of visits to the doctor and dentist, number of pregnancies, etc. during a Specific period of time for each member of the family. A. HOSPital Insurance-~Questions about subscription and coverage of health insurance policies. 5. Health Department--Questions about the knowledge of and use of the facilities of the Lenawee County Health Department. 6. Milk Supply and.Anima1 Diseases--Questions about Bangs Disease, T. B. innoculations for cattle, use Of pasteurized milk, etc. 7. Sanitations~Questions about the source of the family's water supply, method of sewage diaposal, etc. 8. Housing-~Questions about rental or ownership of the reapondent's home. 53 Of these eight areas with which the health section of the question- naire was concerned, more of the interviewers indicated problems with the sanitation question than.with any other specific area. Eight of the 52 interviewers particularly mentioned having difficulty with the questions concerning water supply and sewage disposal. Some of the reapondents were reported to have been Opposed to these questions, be- cause they felt that the source of their water and their method of sewage disposal was their own business. Others feared that the Govern, ment was sponsoring the survey in.order to find out if the respondents were violating the laws that apply to the methods of legal sewage dis- posal. Five interviewers (See Table II) reported having difficulty with the questions having to do with medical care. They said that the re- spondents could not see the purpose of these questions and implied that they were rather personal. By far, the majority of the problems reported on the Health part of the questionnaire were not due to any one question or area but were problems concerning the entire Health part. The respondent in many cases, the interviewers stated, could not be made to understand that the Health Council and Health.Department were interested in having the information about their health problems so that they could do something about them, and that the findings were not to be used by the Federal or State Governments. The respondents were reported to suspect that the survey was being made as a beginning to a socialized medical program. Others of the reapondents thought that a health survey was a waste of Sh TABLE II DISTRIBUTION OF TEL 52 INTERVIEWERS REPORTING-PROBLEMS ON THE HEALTH QUESTION: BY THE AREAS ON WHICH THEY HAD PROBLEMS Areas on the Health __Part of Questionnaire Number of Interviewers l. Immunization (2) XI? 2. Past Illnesses (3) xxx 3. Medical Care (5) XXXXX 14. Hospital Insurance (2) XX 5. County Health Department (1) x 6. Milk Supply anthnimal Diseases (2) IX 7. Sanitation (8) xxxxxxxx 8. Housing 9. Health.Questions in General Total 52 *Legend: Each X represents one interviewer time and money, because Lenawee County did not have any health problems. As can'be seen from Table II, none of the other areas of the Health part of the questionnaire accounted for more than two or three of the problems. B. Types of Problems Reported by the Interviewers It was possible through the use of one of the questions asked of the volunteer interviewers by members of the project staff studying the SS organization of the Lenawee County survey ('On which questions did you find it most difficult to get answer? Why did people find these ques- tions difficult? Item lh,.Appendix:B) to develop a typology of inter- viewer problems. The reasons for interviewer difficulty given in answer to the last part of the question seemed to pattern themselves around two categories: problems of ccmmunication.and problems of op- position. These are the two types of problems which will be used for statistical analysis in this paper.“ They are defined as follows: 1. Problems of Communication--the reapondent could not be made to understand the question or the purpose of the survey; The interviewer was not able to answer the questions asked him concerning either a particular item or the survey in general. 2. Problems of Qpposition-—the respondent was opposed to an area of the questionnaire or to the survey in general. Unspecified or general Opposition, Opposition on the basis of a fear of socialized medicine or governmental inter- ference, and Opposition to the questionnaires as being too personal or a waster of time, were reported. Some of the statements which form the basis for deriving these categories are given below. .After each statement, the type of problem which it represents is identified, and an explanation of this identifi- cation is given: 56 Sometimes on those questions about housing peOple; well, peOple are afraid they are obligated to take peOple in.“ (Communication--the interviewer did not make it clear to the reapondent that the survey was only intended to get an estimate of how many evacuees Lenawee County could care for easily in case of an emergency and was not meant to hold anyone reSponsible for a Specified number.) Housing‘Atomic Bomb Evacuees--one family didn't want to stick its neck out. They thought the Government would be sending them someone to take care of right away. (Com- munications—same reason as above.) People's illnesses-~thought they were nobody4s business. (Opposition-—the basis for this Opposition is the reapond- ent's consideration that his medical history is his own private record.) Whether used County Health.De artment Services--It was none of my (the interviewer's business. (Oppositionr- same reason as above.) One Home Economist training-éDid high school home economics courses count? I (the interviewer) didn't know. Mr. (...) said something about high school training. I think he said it didn't count. I couldn't explain to them why we wanted to know whether they owned their own hmme, or about the septic tank. (Communications-the interviewer did not know the answer to the questions asked her by the reapondent, and therefore could neither explain the purposes nor the limits of the questions.) Whole health.part-4They (the reapondents) answered as if they would tolerate the questions but didn't like to; all the questions pertaining to health were considered foolish and of no use. And.besides, peOple felt that it was none of the Government's business. (Opposition on the basis of governmental interference in local health affairs and op- position to the health survey as a waste of time. These are typical example of the statements recorded on the inter- view schedules. These have been coded either as problems of communi- cation or problems of Opposition. The following distribution indicates the extent of each of these types of problems: 57 TABLE III DISTRIBUTION OF INTERVIENERS BI THL;TYPES OF PROBLEMS WHICH THEY REPORTED ON THE ENTIRE QUESTIONNAIRE Types of Problems Interviewers By Percent 1. Problems of Communication (lbs) XXXXXXXXXXXDCX 2. Problems of Opposition (16%) W 3. No Problems Reported (70%) MW Total 100% Number of Cases: 313 Legend: Each.X represents one percent. From Table Ill, it can'be observed that lb% of the interviewers stated that they encountered.problems of communication, and 18% of the interviewers stated that they encountered problems of Opposition. These percentages represent us to h? interviewers, reapectively. There were 215 interviewers who said that they encountered no problems at all, and four that did not reSpond to the question. These numbers, rather than the percentages given in Table III, will be used in the Chi-Square con- tingency tables in the coming chapters. The total for each table should be 309 unless Specified otherwise. C. Limitations of This Method for Defining Interviewer Problems There are three rather specific reasons why the data concerning interviewer problems may be rather incomplete. All of these limitations 58 are a function of their being reported interviewer problems rather than actual interviewer problems. The first difficulty in using reported problems of the interviewers is that, as volunteers with little previous experience as interviewers, they probably did not recognize all of the problems which they encountered and therefore could not report them. The second limitation is that there was a delay for some of the interviewers between the time they completed their interviewing and the time when they were contacted concerning their difficulties. This was due to the fact that the research team did not attempt to enter a town- ship until their survey was completed. Therefore, even if the interview- ers had encountered and recognized problems, they might not have re- membered them. It is considered, however, that at least the 9h inter- viewers who reported problems actually encountered them. The third limitation is one which could equally be a function of the interviewer's training. It is known that at least the interviewers who attended the meetings instructed by Michigan State College personnel were told about the problems which they could expect to encounter while interviewing. .Also, they had an.0pportunity to see how some of these problems could be overcome if they watched the demonstration interviews. There is a possibility that the threshold for their recognizing and re- porting problems of communication and Opposition was heightened by the training. All of the above limitations should be taken into account by the reader in interpreting the findings. CHAPTER V TESTS FOR VARIABLES RELATED TO INTERVIEWER PROBLEMS This chapter is concerned with the factors which have been suggested might be related to the reported.problems of the volunteer interviewers in the Lenawee County survey; The factors will be broken down into three categories as they were in.Chapter I where the hypotheses were stated. These categories are: Social Characteristics of the Interviewh ers; Previous Experiences of the Interviewers; Involvement and Partici- pation of the Interviewers in the Survey; As pointed out earlier, the Chi-Square statistical method will be used in this chapter to determine whether the hypothesized relationships actually exist. A. Social Characteristics of the Interviewers It was indicated in.Chapter I that certain social characteristics of the volunteer interviewers might be related to the reported problems of these volunteers. Positive relationships between sex or age and problems of communication and apposition were hypothesized; likewise, it was hypothesized that there would be no relationship found between edu- cation or place of residence in Lenawee County and the above named inter- viewer problems. The distribution of each of these variables by the types of interviewer problems appear below, as does the Chi—Square ob- tained through use of the method described in Chapter III. Sex of the Interviewers The interviewing was done primarily by women. 0f the 313 inter- viewers, 293, or 9h%, were women, while only 20, or 6%, were men. Table IV below is designed to test the following hypothesis concerning the sex Of the volunteer interviewers: Men.report_problems of Opposition and communication.more frequently than do women. TABLE IV DISTRIBUTION OF TYPES OF INTERVIEWER PROBLEMS BY THE SEX OF THE INTERVIEWER Types of Sex of the Interviewer Interviewer Problems Male Female Total Communication if hh hS Opposition IQ DB h9 NO Problems 12 203 215 Total 19 290 309w 12 - 2.85 P - .20 - .30 *‘In several cases in this thesis, it was found that the expected fre- quency was less than five. These will be indicated by underlining and Observed frequency. The computation of the 13 in these instances has necessitated that a correction factor Of .5 be subtracted from the difference between the Observed and the expected frequency in each cell. This correction factor was suggested by Donald Lewis and C. J. Burke, "The Use and Misuse of the Chi-Square Test" thOIOgical Bulletin, h6(l9h9), pp. h33-h89. There were four interviewers who were not asked or did not reapond to the questions concerning interviewer problems. One of these was a male; the other three were females. Reference will be made to only the totals which.are different than 309 from now on. J l .-' I l - 61 A Chi-Square with two degrees of freedom yields a probability value of .20 - .30. This means that the hypothesis that men encounter prob- lems of Opposition and communication more frequently than do women can be rejected. It appears that neither sex encountered significantly more K; problems than did the other. It is, of course, true that the sample of ZFmen was very small and that some criticism can be levied at any attempt i, to obtain a Chi-Square value at all. By using the sign-test suggested by Hagood, one can determine what direction the association would take had a significant association been revealed. The number of problems Of Opposition reported by the male interviewers was greater than the. expected frequency, while the number of problems of communication reported by the female interviewers was greater than expected mathematically. It should be remembered that these differences are not significant and only indicate tendencies toward one direction or another . Age of the Interviewer The age of the interviewers ranged from 19 years to more than 70 years. However, the maiority of the interviewers'were young, middle- aged women. The Median age was thirty-eight years. The distribution of ages by percentages is included below to provide the reader with a more precise picture of the interviewers' ages: TABLE V DISTRIBUTION OF THE INTERVIEW'ERS BY AGE Age Percent 1. Below thirty years 2 . Thirty to thirty-nine years 3. Forty to forty-nine years )4. Fifty years or Older Total Number of Cases: 311* Iagend: Each 1 represents 1% * Two of the interviewers did not provide their age. The follovdng hypothesis concerning age will be tested: Interviewers who are fifty years or older report problems of communication and Opposition more frequently than do those who \_ x are less than fifty years Old. 63 TABLE VI DISTRIBUTION OF TYPES OF INTERVIEWER PROBLEMS BY THE AGE OF THE INTERVIJMFRS Type Of Age of the Interviewer Interviewer Problems Below Eifty & Total Fifty Above Communication 39 6 hS Opposition NB 6 h9 NO Problems 171 , h2 213 Total 253 5b 307* 1:2 - 2.21 P - .30 - .50 * There were two interviewers who did not answer the question COD! cerning their age, and there were four interviewers, as previously mentioned, who did not answer the questions concerning interviewer problems. A Chi-Square Of 2.21 was obtained. ‘With two degrees of freedom, this Chi-Square value could be expected to occur by chance between thirty and fifty times in every hundred distributions. Therefore, the hypothe- sis that age is related to problems of communication and Opposition in the manner suggested above should be rejected. It is interesting to Observe, however, that even though the differ- ence is not significant, that the direction of association would have been toward the reverse of what was hypothesized. That is, the inter- viewers who were fifty years or Older reported fewer problems either Of communication or Opposition than expected by chance, but not significantly 614 fewer to establish any conclusions about this tendency. If one were i to chance a Speculation, however, he might suspect that the Older peOple were better known.and therefore encountered less resistance. Education of the Interviewers The amount of education experienced by the interviewers varied greatly; Twentyhsix of the participants had less than an eighth grade education. On the other extreme, twenty-three of the interviewers were college graduates. Three-quarters of the interviewers had at least a high school education. The distribution of the interviewers' education by percent is given below: TABLE VII DISTRIBUTION OF INTERVIEWERS BY EDUCATION Last School Year Completed Percent 1. Eighth Grade or Less (8%) xmxm 2. Some High School to High School Graduate 3. Some College or Business School(25%) Total 100% Number Of'Cases: 313 Legend: Each X represents 1% Table VII is designed to test the following hypothesis: 65 There is no difference among the interviewers,possessing differ- ent educational levels in the extent to which they encounter pgoblems of communication.and Opposition. TABLE VIll DISTRIBUTION OF TYPES OF INTERVIEWER PROBLEMS BY THE EDUCATION OF THE INTERVIEWERS Type Of Education of the Intervieyggs Interviewer Problem Eighth Grade ’SOme High Some College Total or Less §9h091;_v Experience Communication ‘5 25 15 h5 Opposition 6 3h 9 h9 No Problems 15 lh8 52 215 Total 26 207 76 309 x2 - 3.36 P - .30 - .50 .According to Fisher and Yates' Probability Table, a Chi-Square of 3.36 with two degrees of freedom is equal to a probability of between .30 and .50. This means that such a distribution Of interviewer problems by education could occur by chance 30 to 50 times out of every hundred. This is too great a chance factor to permit one to conclude that there is a relationship between the two variables. 'When the signs-test is applied to this data, there is not even a consistent direction of associ- ' ation indicated. For these reasons, the hypothesis positing no relation-' ship between interviewer problems and education can be accepted. 66 One might suggest that the interviewers who had experienced onLy eight years of formal education or less might not have recognized a problem when they encountered it. If this were the case, at least with the signs-test, some direction to the distribution might have been recognized. That is, the interviewers with.the least education would be expected to also have fewer interviewer problems. There is no indi- cation Of this either. Location Of Residence in Lenawee County The interviewers were not solely from small towns or farms as might be expected, when one considers that the interviewing being studied in this thesis was carried out in the areas Of the county outside the county- seat city. _Actually, there were some of the peOple from the city of Adrian.who interviewed in the rural survey. Of the 313 interviewers, 79, or 25%, were from small cities, such as Blissfield, Clinton, Hudson, Morenci, Tecumseh, or.Adrian. Since the survey had a rural emphasis, it was felt that some of the interviewers from the small cities might have encountered difficulty with the questions on the survey form which.were directed toward farm life. The following hypothesis will be tested to determine if the lo- cation of the interviewer's residence is related to his reported inter- viewer problems: Interviewers from the small cities report problems Of communication and Opposition.more frequently than do those from the rural areas. The x2 or 2.09 obtained in Table II is significant at only the .30 to .50 level. This indicates that such a distribution could have 67 occurred by chance thirty to fifty times in every hundred, and that therefore, there is no relationship between the variables. The signs- test, likewise, indicates no definite patterning of the distribution of interviewer problems by the location Of the interviewer's home in Lenawee County. TABLE II DISTRIBUTION OF INTERVIEWER PROBLEMS BY THE LOCATION OF‘THE INTERVIEWER'S RESIDENCE 1N LENAWEE.COUNTY Types of Location of Interviewer's Residence Interviewer Problems Small Town Small Total and Open Country Cities Communication 31 1h h5 Opposition hO I 9 N9 NO Problems 159 56 215 Total 230 79 309 I“ - 2.09 P - .30 - .50 The tables in this section have shown that none of the social characteristics Of the interviewers are Significantly related to inter- viewer problems. Some of the tables indicated that there was a tend- ency toward association in some cases. For example, in the case of age, those interviewers who were more than fifty years old reported fewer problems than expected, but not few enough to be significant. B. Previous Experiences of the Interviewers It was suggested in.Chapter I that certain of the previous experi- ences Of the interviewers might have provided them with information, greater confidence, techniques for resisting Opposition, etc. which would be of use in interviewing for the self-survey. Particularly was it pointed out that certain Of these experiences might have helped the interviewers in cutting down the number of problems which they reported. .A problem which was easily overcome might, then, not be considered imp portant enough to mention.or might not'be considered a problem because it was easily overcome. In this section Of Chapter V, certain Of the experiences Of the interviewers will be described, and a Chi-Square test will be used to determine if there is actually a relationship between.any of these experiences and interviewer problems. The experiences which will be considered are: previous sales experi- ence, previous interviewing experience, occupational experience, organiza- tional membership, officership experience, and previous participation in community-activities. Previous Interviewinngxperience Approximately half of the interviewers had had previous door-to-door interviewing experience. Some had participated in school, church, or the United States Census. Others had solicited funds for Polio, Red Cross, and Cancer Drives. It was suggested that some Of these experiences would have provided the interviewer with techniques for overcoming Opposition, 69 confidence in beginning the interviewing, and perhaps facilitated one's communication with the informant. The hypothesis which will be tested concerning interviewing experience follOws: Interviewers who have had previous experience interviewigg _§eport_pr0blems of communication and opposition less fre- quently than do those who have had no such experience. TABLE I DISTRIBUTION OF INTERVIEWER PROBLEMS.BY PREVIOUS INTERVIEWING EXPERIENCE OF THE INTERVIEWER Types of Previous Interviewing Experience? Interviewer Problems NO Interviewing Previous Inter- Total Experience viewing Experience Communication 20 ' 25 h5 Opposition 21 28 h9 No Problems 116 99 215 Total 157 152 309 x‘3 - 2.79 P - .20 - .30 .A Chi-Swuare of 2.79 yields a probability value Of between .20 and .30. This means that such a distribution as that Of interviewer prOblems by previous interviewing experience could have occurred by chance as many as thirty times out Of every hundred distributions. For this reason, the hypothesis positing a relationship between interviewer problems and interviewing experience will have to be rejected. It is further 70 interesting to note, that although the two variables are not related, the direction of the association would have been the reverse of that hypothesized. In other words, the interviewers who had had previous experience in interviewing reported more problems than would theo- retically have been expected, according to the signs-test. Had the Chi- Square been significant, then it could have been concluded that previous interviewing experience increased the likelihood of an interviewer en- countering interviewer problems. Previous Sales Emperience Only twentybeight of the interviewers reported previous door-to-door sales experience. This represents less than ten percent Of the inter- viewers. It was felt, however, that the Chi-Square method should be applied to the distribution of interviewer problems by previous sales experience to determine if the same situation.as found above in the case of previous interviewing experience also existed in the case Of previous sales experience. The hypothesis, which was suggested in.Chapter I, positing a relationship between door-tO-door selling experience and inter- viewer problems, finds less support now that the hypothesis concerning interviewing experience has been rejected. Table II presents the distri- bution of interviewer problems by previous sales experience, and the Chi- Square obtained. A Chi-Square Of .11 for a contingency table with two degrees of freedom yields a probability Of .90 - .95 when compared with Fisher and Yates Probability Table. This means that the hypothesis that there is 71 a relationship between previous door-tO-door selling experience and interviewer problems can be rejected; the two variables are apparently independent. The signs-test, as would be expected by the high chance factor Of such.a distribution, demonstrates no pattern among the data. TABLE II DISTRIBUTION OF INTERVIEWER PROBLEMS BY THE PREVIOUS DOORJTO-DOOR SALES EXPERIENCE OF THE INTERVIEWERS Types Of Extent of Previous Sales Experience Interviewer Problems No Sales Previous Total Experience Sales Egperience Communication 39 2 h2 Opposition to ‘5 NS No Problems 178 . 20 198 *- Total 257 28 285 x2 - .11 P - .90 - .95 § Twenty-four of the interviewers did not reSpond to the question on their previous sales experience. Previous and Present chupational Experiences Most Of the interviewers, being married women, were, at the time of the survey, engaged in housekeeping as their main occupation. .All inter- viewers who had ever been employed, were included in the categoryx ”previously employed", because it is with the factor of employment experience which this section of the thesis is concerned. 72 The interviewers compose a vast range of employment experiences, and this fact adds weight to the hypothesis that there is no difference between an employed group and an unemployed.group as to interviewer problems, because there is probably little common experience between these groups which would contribute to a difference. The hypothesis is tested in Table II. There were not enough individuals in any one occu- pational group to permit analysis by occupational groups with the exr ception of professionals. These are treated separately in Table XIII. There is more reason than sheer numbers for analyzing the data concerning professional experience separately; Many of the writers in the litera- ture emphasized the fact that those individuals who have had professional trainingtand experience make the best interviewers, particularly those who have a keen interest in science. The main difficulty in making the statistical test is that any relationship Obtained might be a function of their interests rather than their experience.1 The Chi-Square obtained in Table XII is .81. According to Fisher and Yates Probability Table, such a Chi-Square would occur by chance fifty to seventy times out of every hundred. This amount of chance Operat- ing makes it impossible to conclude that the two variables are related. Thus, the hypothesis that previous employment and interviewer problems are not associated can be accepted. 1. This suggestion is made in the following.references: Lundberg, gp. 2251., p. 353; Radyanvi, pp. 939., p. 22; and Guest, gp. 233., p. 279, 73 TABLE III DISTRIBUTION OF TYPES OF INTERVIEWER PROBLEMS BY PREVIOUS EMPLOYMENT EXPERIENCE OF THE INTERVIEWERS Types of -;Employment of Interviewers Interviewer Problems Presently or Pre- Never Total viously Employed Employed Communication 20 25 h5 Opposition 2h 25 h9 NO Problems 89 123 212 Total 133 173 306* x2 - .81 P . .50 - .70 g *There were three interviewers who did not answer the questions concerning their occupational experience. Four interviewers did not answer the question about interviewer problems. A Chi-Square of 1.3h was obtained in Table XIII, when the Chi- Square statistical method.was applied to the data there. From Fisher and'Yates Probability Table, it can be Observed that such a Chi-Square has a probability of .50 to .70. This means that more than one out of every two times, such.a distribution could have occurred by chance. Therefore, the two variables are probably not related. If it is the interest of professionals which contributed to their ability to interview, it has not been shown here. Some other test for Occupational interest, such as the Strong Vocational Interest Test, might serve as a better criterion for understanding the professional as an interviewer. It would seem that it is not his ability to overcome interviewer problems which is the discriminating factor. 7h TABLE XIII DISTRIBUTION OF TYPES OF INTERVIEWER PROBLEMS BY PROFESSIONAL EXPERIENCE Types of Professional Never Interviewer Problems Employment Employed Total Communication 13 25 38 Opposition 9 25 3h NO Problems V hl 123 16h *- Total 61; 173 236 X2 8 1 03h P = 050 "' s70 *- 69 of the interviewers have previously or are at present engaged in some occupation other than a professional one. OrganizationaliMembegship It was expected that peOple who were involved in a self-survey would also be active in other community activities. One indication of this activity should be membership in community organizations. ‘Well over half of the interviewers were found to belong to more than one organization besides the Church, and some of the interviewers belonged to as many as six or seven organizations. The distribution Of number of organizational affiliations is given below; In Chapter I it was suggested that there were experiences which one might have had as a member of an organization which would contribute to his ability to interview. Likewise, it was hypothesized that these experi- ences would be reflected in the interviewer's ability to overcome problems 75 TABLE XIV DISTRIBUTION OF INTERVIEWERS BY NUMBER 0r ORGANIZATIONAL AFFILIATIONS (OTHER THAN CHURCH) Organizational Affiliation Percent 1. NO Organizational affiliation (lh%) xxxxxxxxxxxxxx 2. Membership in one organization (2h%) xxxxxxxxxxxxxxxxxxxxxxxx 3. Membership in two organizations (2h%) xxxxxxxxxxxxxxxxxxxxxxxx h. Membership in three or more (38%) xxxxxxxxxxxxxxxxxxxxxxxxxxxx organizations XXXXXXXXXX Tomi 1mm Number Of Cases: 309 Legend: Each X represents 1% of communication and Opposition. Table XV presents the distribution of interviewer problems by the organizational experience Of the interviewer with the resulting Chi-Square. A Chi-Square of 3.03 was obtained in.Table XV. According to Fisher and Yates Table, such a Chi-Square value would occur by chance between eighty and ninety times out of a hundred. It is therefore necessary to reject the hypothesis that organizational membership experience alone would account for the interviewer's ability to overcome interviewer problems. However, there is another possibility which has to do with the organization.membership Of the interviewers. Perhaps it is only the factor Of officership or other leadership in an.organization which would 76 TABLE XV DISTRIBUTION OF TYPES OF INTERVIEWER PROBLEMS BY MEMBERSHIP EXPERIENCE IN COMMUNITY ORGANIZATIONS OTHER THAN CHURCH Types of Number of Interviewer Problems Organizational Affiliations Total None One Two Three or More Communication A 12 9 20 NS Opposition 8 9 12 20 A9 NO Problems 30 5h 53 78 215 Total 12 75 “it 118 309 X2 8 3.03 P = .80 - .90 provide the interviewer with the skills required in overcoming problems Of communication and Opposition. If this were true, then the inter- viewers who were Officers of these community organizations might en- counter significantly fewer interviewer problems than those who were not or never had been in these positions. It was hOped that this hypothesis might be tested, but there was no infOrmation available for analysis concerning it. Participation in Community Activities It was suggested in Chapter I that for two reasons the interviewers who have previously participated in community activities might be exr pected to encounter fewer problems Of communication and Opposition. The first reason was that the interviewers might well have acquired skills 77 in this experience, esPecially in selling a particular program to the community, which would transfer to the interviewing experience.“ Like- wise, the second reason was that certain of the interviewers might be recognized for their participation in previous programs, and because of being recognized be received for the interview'more readily by the informants. Table XVI is designed to test the following hypothesis: Interviewers who have had previous experience in_participating in community activities encounter fewer problems of communica- tion and opposition than those interviewers who have not had this experience. TABLE XVI DISTRIBUTION OF INTERVIEWER PROBLEMS BY PREVIOUS PARTICIPATION IN COMMUNITY.ACTIVITIBS Types of No Previous Previous Interviewer Problems Participation Participation Total Communication 17 28 hS Opposition 19 30 E9 No Problems . 91 12h 215 Total 127 182 309 x2 = .15 P = .70 - .80 A Chi-Square of .hS was obtained in Table XVI. The probability that such a distribution could have occurred by chance is between .70 and .80. Consequently, it is necessary to reject the hypothesis that 78 interviewers who participate in community activities encounter fewer interviewer problems than those interviewers who did not. Apparently, none of the previous experiences of the interviewers had any carry-over effect in terms of helping them overcome interviewer problems. C. Involvement and Participation of the Interviewers in the Survey This section of Chapter V is concerned with the type of organizational method used in each township for getting the volunteers interested in interviewing for the self-survey. It is also concerned with the number of -interviews the volunteer obtained, the degree to which the reSpondent was aware of the self-survey, and the training which the interviewer experi- enced before he began the interviewing jab. Each of these variables will be considered from the standpoint of their possible relationship to interviewer problems . Method of Township Organization By “method of township organization" is meant primarily the lines which the township organization used for involving the interviewers in the survey. These have been outlined in detail in.Chapter I (p. 3). It was suggested at the time that different types of obligations are inp voked in the use of each of these lines of involvement, and that these might be in some way related to the extent to which the interviewers would report encountering interviewing problems. The specific hypothesis reads: 79 There is a difference among_distributions of interviewers by the method of township organization in the extent to which they report problems of opposition and communication. When the Chi-Square value of 6.hS, obtained in Table XVII, is compared with.Fisher and Yates Table of Probability, it is found that such a distribution could be expected to occur by chance between ten and twenty times out of every hundred. Although this does approach.the five percent level which was selected as the point at which hypothesis would be accepted or rejected, it is not close enough to the five per- cent level to say that the hypothesis could be accepted. It is interesting, however, to note the direction in which this distribution approaches association. Those interviewers who participated in surveys where the formal lines of theschool district were utilized reported more problems than would theoretically be expected, while those who participated in surveys where friendship lines were utilized reported fewer problems of either communication or Opposition than theoretically expected. Those who participated in surveys where association lines were utilized reported approximately the number of problems theoretically expected. It must be remembered, however, that association has not been shown here, and that tendencies toward association in a particular direction cannot be interpreted as very decisive. Structural Position of the Interviewer The organizers of the survey in the local areas had an Opportunity to learn.more details about the survey, to have the experience of selling 80 TABLE XVII DISTRIBUTION OF INTERVIEWER PROBLEMS BY METHOD OF TOWNSHIP ORGANIZATION Types of _;_ Method of Township Organization Interviewer Problems Friendship Kssociational School-District Total Lines Lines Lines Communication 5 18 22 hS Opposition 7 22 2O h9 No PrOblems 52 88 75 215 Total oh 128 117 309 x3 - 6.16 P - .10 - .20 the idea of the survey to the other interviewers, to meet the county leaders and thus see the survey on its larger base, etc. As a result of these experiences, it was hypothesized that these organizers might have less difficulty interviewing, measured in terms of their reported interviewer problems. Table XVIII was set up to test this hypothesis. The Chi-Square obtained from the application of that statistical method to the data in Table XVIII was 3.67. The P-Value which results when such a Chi-Square is compared with Yates and Fishers Probability Table is between .30 and .50. This indicates that such a distribution as the above could have occurred by chance as many as fifty times out of every hundred. Therefore, it is necessary to reject the hypothesis that the interviewers who were also organizers encountered significantly fewer problems, because they did not. 'With only 21 organizers, it is 81 TABLE XVIII DISTRIBUTION OF INTERVIIMER PROBLEMS BY THE STRUCTURAL POSITION OF THE INTERVIEWER IN THE SURVEY ORGANIZATION Types of Structural Position of the Interviewers Interviewer Problems Organized and Interviewed Total Interviewed Only Communication ‘2 h3 h5 Opposition l_ h2 h9 No Problems 12 203 215 Total 21 288 309 12 - 3.67 P = .30 - .50 difficult to recognize any tendencies for the data to align themselves in.any particular direction. There is some indication, however, that the organizers did encounter fewer problems of communication than theo- retically expected. There is some logic to this tendency, for these interviewers should have known more about the content of the survey be- cause of their part in organizing, and would be expected to be able to communicate it to others as they did to the volunteers whom they con- tacted. Interviewer Training There are different types of training which the volunteer interviewers associated with the Lenawee County Health and Defense Survey might have experienced. These will be considered in greater detail in.Chapter VI 82 where the question of training is the primary concern. Here, the attention will be paid to the created dichotomy; interviewer training vs. no interviewer training. It was pointed out in Chapter I that the agencies which employ interviewers pay particular attention to the factor of interviewer training, and that there is a great amount of concern with it in the literature. One study was reported in.which a relation- ship was observed between the number of problems which a group of inter- viewers reported and the amount of training that they had received.2 A hypothesis has been derived based on that finding. It will be tested in Table XIX. Interviewers who have not been trained report pgoblems of com- munication and opposition.more frequently than do those who have been trained. The Chi-Square of 21.71 obtained in.Table XIX is higher than the 13.815 needed to obtain a probability value of .001. This means that there is much less than one chance in.a thousand that the distribution could have occurred by chance, and that, therefore, interviewer problems are associated with interviewer training. The corrected coefficient of mean square contingency'(5'==.h3) indicates a moderate degree of association, 'With sociological data, and eSpecially when there are as many as 309 cases, this is a satisfactorily high degree of association to add weight to the finding that the two variables are associated. 2. Knower, ER. 933., p. 223. 83 TABLE XIX DISTRIBUTION OF INTERVIEWER PROBLEMS BY INTERVIEWER TRAINING Types of Extent of Training_ Interviewer Problems Non-Trained Trained Total Communication 18 27 hS Opposition 17 32 h9 No Problems 30 185 215 Total 65 2th 309 x2 - 21.71 0' =- .h3 P .001 The direction of association is as hypothesized. Those interviewers who received interviewer training encountered significantly fewer prob- lems than those who were not trained. This finding will be explored in greater detail in Chapter VI when training is discussed exclusively. Number of Interviews Completed As pointed out in an.ear1ier chapter, there is a possibility that most of the interviewers had difficulty with their first few interviews, but as the job became more familiar, forgot the difficulty which they had with the first few interviews. It was also pointed out that there is a possibility that the more interviews which the interviewer completed the greater was his chance of calling upon someone who was Opposed to the whole idea of the survey. To determine if either of these could account for the occurrence of interview problems, a distribution of interviewer 8h problems by the number of interviews completed.by the volunteer was made and can be seen in Table XX. TABLE XX DISTRIBUTION OF INTERVIEWER PROBLEMS BY THE NUMBER OF INTERVIEWS COMPLETED BY THE INTERVIEWER V~ Types of _Number of Interviews Completed ’ Interviewer Problems One to Ten to Twenty and Total Nine Nineteen .Above Communication 11 2h 10 h5 Opposition 12 15 22 h9 No Problems 60 80 75 215 Total 83 119 _ 107 309 x8 = 7.29 P . .10 - .20 A Chi-Square of 7.29 was obtained in the above table. .According to Yates and.Fisher's Table of Probability, such a ChiQSquare with four degrees of freedom yields a P-Value of between .10 and .20. This means that such a distribution could have occurred.by chance as many as twenty times in every hundred, and that the hypothesis positing a relationship between interviewer problems and the number of interviews completed must be rejected. The signs-test when applied to the above distribution gives some indication that the more interviews completed, the greater the chance for the interviewer to encounter problems. There are fewer interviewers 85 than theoretically expected who completed between one and nine inter- views and encountered either problems of communication or Opposition, and there are more interviewers than expected who completed more than twenty interviews and encountered problems of Opposition. The differ- ences between the Observed and theoretical frequencies in these cases were not large enough to permit a significant Chi-Square, however. Extent to Which the Respondents Had Heard About the Survey Many Of the peOple in Lenawee County had heard about the survey .before they were contacted for an interview; The Adrian Daily Teleggam had reported the completion of the Rollin.Township self-survey of health, and mentioned that most Of the other townships would be planning and conducting their own surveys within the next few months. When the interviewers visited a household, they probably were able to determine from the informant's response to them whether he had heard about the survey before the interview. .Among the questions on the Social Research Service interview schedule was one asking the informants to estimate what percent of the peOple had heard about the survey. Since most of the publicity given the health survey was favorable, it is expected that those who had heard about it would have been.more favor- ably diSposed to it and also more familiar with the content of the questionnaire than those who had not. If an interviewer contacted peOple who had heard about the survey, it would seem then that he should have less difficulty explaining it to them and likewise should encounter less Opposition. The following hypothesis will be tested to determine if this was the case: 86 There is a difference among distributions of interviewers by the extent to which the reSpondents had heard about the survey in the frequency with which they reported problems of communication and Opposition. TABLE XXI DISTRIBUTION OF INTERVIEWER PROBLEMS BY THE EXTENT TO WHICH THE INFORMANTS HAD HEARD.ABOUT THE SURVEY BEFORE BEING CONTACTED FOR.AN INTERVIEW Types of Householders Who Had Heard About Survey Interviewer Problems None or Few Some Had Most or.All Total Had Heard Heard Had Heard About Survey- About Suryey' About Survey Communication 17 13 ' l2 h2 Opposition 18 ~19 12 h9 No Problems 77 59 70 - 206 5 Total - 112 91 9h 29? I2 - 2.70 P - .50 - .70 * There were sixteen interviewers who did not answer this question. A Chi-Square of 2.70 was obtained in.Table XXI. The P-Value which was determined from Yates and Fisher's Table was .50 to .70. This means that such a distribution as the above could have occurred by chance more than once out of every two distributions. There is not reason to accept the hypothesis that interviewer problems are related to the extent to which the informants knew about the survey before they were called upon 87 for an interview. This finding has implications for the researchers who purport that the pOpulace should know about the survey before it is carried out. In terms of problems for the interviewer, apparently there is no difference effected if the peOple have'been informed about the survey or not. 314mm Chapter V has been a very inclusive chapter. Within.it there have been at least fifteen hypotheses subjected to the Chi-Square method of statistical analysis, and some interesting results have been forthcoming. {‘It was found that none of the social characteristics which are Often required by interviewer agencies accounted for a significantiy large enough number of the interviewer problems to attribute to them a causal effect. Likewise, it was found that none of the previous experiences of the interviewers were transferred to the interviewer situation so as to be reflected in the extent to which they encountered interviewer problems . Similarly, the method of involvement Of the interviewers, although indicating some tendency for interviewers who were involved by friends to have encountered fewer interviewer problems, could not account for all of the interviewer problems. This is true of the extent to which the informants had heard about the survey before being contacted for an interview and is also true Of the number of interviews the volune teer completed. None of these variables bore an association with the variable of interviewer problems. The one variable which showed a positive association was that Of interviewer training. It appears that 88 the interviewers who were not trained encountered significantly'more problems than those who were trained. Because of the relatively high degree of association found, it is considered necessary to consider the variable Of interviewer training more carefully. This will be done in Chapter VI. CHAPTER VI TRAINING OF THE INTERVIEWERS In the previous chapter, it was found that the only variable analyzed in this study so far which is significantly associated to the problems of the interviewers is that of interviewer training. Because of this, the interviewer's training will be analyzed in greater detail to determine if there are any elements within the training process which.might account for some Of the volunteers reporting interviewer problems and the others not reporting interviewer problems. Various types of training and aSpects of the training programs will be discuss- ed below: A. Types of Interviewer Training There are different types of training which the volunteer inter- viewers associated with the Lenawee County Health and Defense Survey might have experienced. These include: 1. Instruction.by Michigan State College personnel at a regular training meeting. 2. Instruction by'a local organizer at a township or village training meeting not attended by Michigan State College personnel. 3. Individual instruction by another member Of the survey organization. 9O gaining Meetings Attended by Michigan State College Personnel There were eight training meetings which were attended by the Health Specialist of the COOperative Extension Service and another representative from Michigan State College. These meetings were held on the following days for the Specified townships and villages: 1. August 13, 1951 for Rollin Township. 2. September 13, 1951 for Woodstock and Cambridge Townships. 3. September 114, 1951 for Morenci Village, Hudson City, and Hudson, Seneca, and Medina Townships. )4. September 214, 1951 for Clinton, Adrian, Rome, Madison, Ogden and Fairfield Townships . 5. September 27, 1951 for Morenci Village. 6. October 2, 1951 for Franklin and Ridgeway Townships. 7. October 11, 1951 for Dover Township. 8. November 9, 1951 for Deerfield Township. At each Of these meetings, one Of the Michigan State College men read through the mimeOgraphed instruction sheet with the interviewers. This instruction sheet included a description of the survey, its pur- poses, some general suggestions concerning interviewing, and the answers to most of the questions which the informants were expected to ask about the survey. Likewise, one of the two men usually read through the sur- vey questionnaire one item at a time, explaining the purpose Of each of the questions and pointing out some of the difficulties which the inter- viewers might expect to encounter on the questions. Frequently, a demonstration skit was acted out by persons who had participated in the Rollin Township survey, or by the men from Michigan State College . 91 Following the demonstration, the volunteers were asked to pair Off and practice interviewing each other under the supervision Of the college men. During these practice interviews, many problems were brought up that the instructors answered at that time. Interviewers who attended this type Of meeting probably received comparable training. Of the interviewers who participated in the survey, 107, or 3h%, received training at meetings instructed by Michigan State College personnel. TrainingLMeetings Not.Attended by Michigen State College Personnel Since neither minutes not descriptions of the meetings not attended by Michigan State College personnel are available, the information about these meetings has been taken from the interviewer's reports of these meetings. In answer to the question, "Can you tell me how you were given instructions to do the interviewing job?” (Item 11, Appendisz), the interviewers supplied brief descriptions Of the training.meetings. From these descriptions, it was Observed that the training differed some- what from township to township when Michigan State College personnel were not present. However, it was Observed that at each Of these meetings there were copies Of the instruction sheets available, and that the inter- viewers either read and discussed these, or the chairman of the meeting read them aloud and discussed them. Since most Of the chairmen had previously attended meetings at which.Miohigan State College personnel were present, they apparently attempted to follow the pattern established at these meetings. For this reason, training at these meetings is accepted as being similar. 92 0f the 313 interviewers who participated in the survey, St, or 17% Of these, experienced training at meetings like the above, where Michigan State College personnel were not present. Individual Training There are many of the interviewers who did not attend any of the training meetings. Most of these, if they were trained, received their instructions from another member of the survey organization who usually had attended one of the meetings. Individual training Of this type varied greatly. Some of the interviewers described their training experiences as simply reading through the mimeographed instructions, while others said that they not only read the instructions, but also went over the questionnaire item by item with their trainer and had a practice inter- view; 0f the 313 interviewers, 87, or 28%, received their training in this way. There were also 6h interviewers, representing 21% of the interviewb ers, who said that they did not experience any type Of training. B. Qpinions About the Sufficiency;of the Trainieg One question which was asked of the volunteer interviewers con- cerned their Opinions of the sufficiency of their training. The Specific question reads: Do you feel that you received sufficient training for the interviewing job you did on this study? (Item 12,.Appendix:B) The distribution Of reSponses shows that by far the majority of the inter- I viewers thought that their training was sufficient. (See Table XXII) 93 There is another question which can be asked about the sufficiency of the training: were the interviewers who attended the training meet- ings the same people who reported that their training was sufficient? A distribution of Opinions about the sufficiency of training by the type Of training actually experienced is presented in Table XXIII. TABLE XXII DISTRIBUTION OF INTERVIEWERS BY OPINIONS.ABOUT THE SUFFICIENCY OF THEIR TRAINING Opinion Percent 1. Decidedly not sufficiently trained (1%) x 2. Not sufficiently trained (11%) xxxxxxxxxxx 3. Uncertain (5%) XXXXX h. Sufficiently trained 5. Decidedly sufficiently trained (6%) xxxxxx Total 100% Number of Cases: 313 Legend: Each X represents one percent The Chi-Square of 22.h2 obtained in Table XXIII is significantly large to conclude that the two variables are associated at less than the .001 level. This means that such a distribution as the above could have occurred by chance less than one time in every thousand. By observing the results Of the sign-test, it can be Observed that those interviewers 9h TABLE XXIII DISTRIBUTION OF THE TYPES OF TRAINING BY OPINIONS ABOUT THE SUFFICIENCY OF THE TRAINING Type of Training Opinions About Sufficiency of Trainingf Total Not Sufficiently Sufficiently Trained Trained Not Trained l7 bl 58 Individually Trained 13 7O 83 Group Trained 8 lh5 153 , a Total 38 250 2911 X2 .. 22.h2 P .001 6 a .hS Those interviewers who reported that they were uncertain concern~ ing the sufficiency Of their training were omitted from the above table. who felt that they were sufficiently trained were the ones who were either trained at a meeting or were individually trained. Those who re- ported that they were not sufficiently trained were in.most cases the ones who were not trained at all or were individually trained. Only eight interviewers who attended training meetings reported that they did not think that they were sufficiently trained. Here, the corrected coefficient of mean square contingency equals .h5. This is a moderately high statistic and indicates that not only were the variables associated, but that the association is moderately high. ‘ 95 C. Types of Training and Reported Interviewer Problems It has been shown in Chapter V that there is an association between interviewer training and interviewer problems. It would seem important now to find out if such a significant association also exists when the trained group is broken down into two categories: individual trained and group trained. The interviewers who were trained individually and those who were not trained at all are compared to that it can be deter- mined if individual training was sufficiently better than no training when interviewer problems are considered; this is done in Table XXIV: TABLE XXIV DISTRIBUTION OF TYPES OF INTERVIEWER PROBLEMS BY INDIVIDUALLY TRAINED AND NON-TRAINED INTERVIEWERS Types of Types of Training Interviewer Problems Not Individually Total Trained Trained Communication 18 1h 32 Opposition 17 15 32 NO Problems 30 57 87 Total 65 86 151 x2 = 6.22 P = .02 - .05 0' = .33 A Chi-Square of 6.22 is Obtained in Table XXIV. According to Fisher and Yates Table, such a distribution as the above would have occurred by chance between two and five times in every hundred. The 96 five percent level of significance has been decided upon in this thesis as the point at which a hypothesis would be rejected. Since the P-value is less than .05, the Hypothesis that interviewers who were individually trained reported fewer problems of communication and Opposition can be accepted. The signs-test likewise demonstrates this to be true. The corrected C-value of .33 indicates that there is a moderate degree of association between the two variables. A similar comparison was made between the non-trained group and those who were trained in group training meetings. A Chi-Square Of 27.59 was Obtained, which is significant at less than the .001 level. A cor- rected C-value of .SN was computed, which shows a slightly higher degree of association between the two variables than in Table XXIV where this value was only .33. The fact that there is a higher Chi-Square value and a higher degree of association Obtained when the non-trained group is compared with the group trained in.meetings than there was between the non-trained group and those trained individually permits one to hypothesize that there is a difference between the group trained in meetings and those trained individually in regard to the frequency with which the interviewers en- countered problems Of communication and Opposition. This hypothesis will be tested in Table XXV. Interestingly, the Chi-Square Of 6.75 Obtained in.Tab1e XXV is significant at less than the .05 level. Noting the signs-test results, one can Observe that the interviewers who were trained in group training meetings encountered significantly fewer problems than did those who 97 were trained individually. The CLvalue Of .26 is, however, rather low indicating a rather low degree of association. TABLE XXV DISTRIBUTION OF INTERVIEWER PROBLEMS BY INDIVIDUALLY AND GROUP TRAINED INTERVIEWERS Types of Types of Training Interviewer Problems Individually Group Total Trained Trained Communication 1h 13 27 Opposition 15 17 32 NO Problems 57 128 185 Total 86 158 2th x3 . 6.75 P =- .02 - .05 E = .26 From Tables XXIV and XXV, it has been determined that those inter- viewers who were trained individually encountered significantly fewer problems than did those who were not trained and that those who were trained in groups encountered problems less frequently than did those who were trained individually, The next problem is to determine if there is a difference between training by professional sociologists from Michigan State College, and training by the local organizers. Table XXVI has been developed for this purpose. The Chi-Square Of .07 Obtained in.Table XXVI has a significance between .95 and .98. It is therefore necessary to reject the hypothesis 98 TABLE XXVI DISTRIBUTION OF INTERVIEWER PROBLEMS BY THE TYPE OF TRAINING MEETING ATTENDED BY THE INTERVIEWER Types of Type of Trainingjfleeting Interviewer Problems Trained by Not Trained by Total , M.S.C. Personnel M.S.C. Personnel Communication 9 h 13 Opposition ll 6 17 No Problems 8h NA 128 Total 10h st 158 I2 - .07 P = .95 - .98 that there is a difference between the two types of training meetings. Local, noneprofessional organizers trained the volunteer interviewers apparently well enough that they did not encounter significantly more problems than those trained by the professional sociologists. This is a significant finding Of this thesis, for it suggests that local organizers can themselves be trained to take the reSponsibility for training the other volunteers, thus making it possible for a self-survey to be organized with less help from outside experts. It was suggested by one of the research team that attention also be paid to the differences between the groups who were trained by local organizers who had attended the meetings headed by Michigan State College personnel and the groups who were trained by local organizers who had not attended these meetings. There were three township meetings where 99 the local organizer had not been trained at a meeting and four where the local organizer had been trained. The comparison is made in Table XXVII: TABLE XXVIJ. DISTRIBUTION OF INTERVIEWER PROBLEMS BY TRAINING MEE'INGS NOT ATTENDED BY MICHIGAN STATE COLLEGE PERSONNEL Types of Types of Training Meetings Interviewer Problems Organizer Did Not Organizer Did Total Attend Meeting Attend Meeting_ Communication ‘ 2 2 h Opposition h 2 6 No Problems 27 17 bh Total 33 21 Sh x2 - .38 P . .80 - .90 It should be recognized first of all in interpreting Table XXVII that the frequencies are almost too small to be used in Chi—Square analyh sis. Even without the correction factor of .5 being subtracted from the 1 differences between the frequencies, the Xz is far from significant. However, it would seem by inspection that there is no difference between the groups trained by an organizer who had attended a meeting and those trainedey an organizer who had not. This finding probably indicates that the organizers who had not attended a meeting had been contacted and informally trained by either Steinicke or Gibson, and thus were prob~~ ably sufficiently'prepared to handle the meeting. 100 D. Aspegts of the Training PrOgram There were certain data obtained from the interviewers concerning ,their training which it is considered may be related to whether they encountered interviewer problems or not. It has been shown above that attendance at a training meeting had a positive effect in cutting down the frequency with which the interviewer encountered.problems. It is hOped here to detenmine some of the elements of the training meetings ‘which.might account for this difference, especially between individual training and group training. For example, 191 of the interviewers reported that they had read the instruction pamphlet, Suggestions for Interviewers. The question which arises quite naturally here is whether these 191 encountered fewer interviewer problems than did those who did not read the instruc- tions. A distribution of these two variables can be found in Table XXVIII. A Chi-Square of hh.03 was obtained when the Chi-Square method was applied to the distribution above. The Chi-Square value was high enough to reject the hypothesis that the two variables are independent at the .001 level of significance. This indicates that the distribution could not have occurred by chance more than one time in every thousand. Like- wise, the corrected C-value was .58 which is relatively high for socio- logical data. Remembering that the corrected C is similar to the Pearsonian r, so that the degree of association must be interpreted as being moderate as would be the case with an r of .58. The signs-test indicates that the interviewers who read the instructions reported 101 TABLE XXVIII DISTRIBUTION OF INTERVIEWER PROBLEMS BY INTERVIEWfltS WHO HRVE OR.HAVE NOT READ‘THE INSTRUCTIONS Types of Reading of the Instructions Interviewer Problems Read Did Not Read Total Instructions Instructions Communication 20 25 hS Opposition 13 36 149 No Problems ‘ 158 57 215 Total 191 118 309 KB . hum P .001 E - .58 significantly fewer problems than those who did not read the instruc- tions. Of the interviewers, 1b? said that they had a practice interview before they began to interview. Like the above, a distribution of inter- viewers who had had a practice interview and interview problems was possible. This can be found in.Tab1e XXIX. A Chi-Square value of 22.51 was obtained in Table XXIX, this value is significant at less than the .001 level. The degree of association was found to be moderate. By observing the direction of the association, it was found that those interviewers who had a practice interview before they began the interviewing job encountered significantly fewer problems than.those who had not had a practice interview. 102 TABLE XXIX DISTRIBUTION OF INTERVIEWER PROBLEMS BY INTERVIEWERS'WHO HAD.AND HAD NOT EXPERIENCED.A PRACTICE INTERVIEW Type of Practice Interviews Interviewer Problems Had Practice Had No Practice Total Interview' Interview Communication 10 35 :5 Opposition 16 33 h9 No Problems 121 9b 215 Total it? 162 309 x2 . 22.51 P .001 6' . .h3 So far, concerning interviewer problems and interviewer training, it has been found that those interviewers who were trained encountered fewer problems than those whO'were not trained, and that those particu- larly who were trained at the meetings least frequently encountered problems of Opposition and communication. Likewise, it has been deter— mined that those interviewers who read the instruction sheet and those interviewers who had a practice interview before they began the inter- viewing job reported problems of communication and Opposition less fre- quently than did those who did neither of these things. There is left but one task, and that is to verify that those interviewers who attended training meetings had a greater chance to experience a practice interview and to read the instruction sheet than those who did not, for this would in part account for the differences between those who were trained 103 individually and those who were trained in meetings. A distribution of type of training by the variable of reading the instructions can be found in Table XXX, and a distribution of type of training by the vari- able of having a practice interview'can be found in Table XXXI. TABLE XXX DISTRIBUTION OF TYPES OF TRAINING-BY INTERVIEWERS WHO HAD READ THE INSTRUCTIONS Type of Reading of the Instructions Interviewer Training Read Did Not Read Total Instructions Instructions Individual Training h3 h3 86 Group Training 1&8 17 162 Total 191 60 2&8 x2 - Sh.2s P .001 E - .66 TABLE XXXI DISTRIBUTION OF TYPES OF TRAININGPBY'INTERVIEWERS WHO HAD PRACTICE INTERVIEWS Types of Practice Interview Interviewer Training Had Practice Had No Practice Total Interview Interview Individual Training 16 7O 86 Group Training 131 31 162 Total lb? 101 2h8 x2 . 90.23 P .001 E a .82 10h A Chi-Square of Sh.2§ was obtained in.Tab1e XXX. This indicates that there is a significant difference between the two types of train- ing prOgrams in the extent to which the interviewers read the mimeo- graphed instructions on interviewing. The difference was significant at less the .001 level. The direction of the association was such that it was demonstrated that those interviewers who were trained in a group training meeting also read the instruction sheet more frequently than those who were individually trained. Similarly, a Chi-Square of 90.23 resulted when the statistical technique was applied to the distribution in.Table XXXI. The Chi-Square value is much higher than.the value necessary to obtain a P-value of .001. Therefore, it can be concluded that the variables are associated in more than a chance manner. That is, significantly more interviewers had practice interviews who were trained in a group meeting than those who were trained individually; A corrected C-value of .82 was obtained in this table. This represents a high degree of association between the two variables. It is probably true that there were other factors in the group ltraining meetings that helped account for the difference in the success of interviewers to overcome interviewer problems. However, the fact that both the reading of the instructions and the practice interview had such a significant negative association with the occurrence of interviewer problems, these two are certainly important differences between individual and group training. 105 Summary This chapter has been concerned with determining the relationship between interviewer training and interviewer problems. It has been shown that not only is there a high degree of association between the type of training which the interciewers received and the frequency with which they reported interviewer problems, but it has also been shown that certain elements contained in the group meetings, such as the read- ing of instructions and practice interviews, were factors which were highly related to interviewer problems. Perhaps the most important finding of this chapter was that there were no differences which were significant between interviewers who were trained in a group meeting by professional sociologists and those inter- viewers who were trained in a group meeting by local organizers. This finding suggests that some of the reSponsibility often alloted by cmm- munities to outside experts can be placed upon trained local citizens. CHAPTER VII CONCLUSIONS This thesis has been concerned with interviewer problems, that is, with the difficulties which volunteer interviewers for a self-survey of health and defense in Lenawee County, Michigan, reported in describ- ing their interviewing experiences. .An.effort has been.made throughout this study to determine whether certain of the social characteristics of these non-professional interviewers, their previous experiences, or their entrance and participation in the survey were the factors associ- ated with their reporting problems of communication and opposition. It was heped that the findings of this study might have implications for other communities planning self-surveys by suggesting certain factors which were found to be related to interviewer problems that they might attempt to control. It is felt that a concern for interviewer problems in the case of self-surveys is sound, for basic to the self-survey idea is the assumption that volunteers who participated in the survey will want to take part in the action program which is expected to follow the completion of the survey. It would seem logical, although it has not as yet been tested, that the volunteers who encounter an extensive amount of difficulty in the interviewing process would be less inclined to participate in the action prOgram after the fact—finding was completed. 107 It was also hOped that public Opinion and attitude researchers, who are continually striving to improve interviewing techniques, might recognize interviewer problems as a fruitful object for their attention which they have not explored to any great extent. The method used in this study might be applicable to their research. A series of hypotheses were outlined in Chapter I which suggested certain variables which were expected to be associated to interviewer problems. These variables were grouped into three categories: Social Characteristics of the Interviewers; Previous Experiences of the Inter- viewers; Involvement and Participation of the Interviewer in the Survey. The particular types of interviewer problems which were included in this study were those of communication and opposition, ‘When.an interviewer reported that he had had difficulty explaining the purposes of the survey or the meaning of certain items on the questionnaire to the reSpondent, this difficulty was designated as a problem of communi- ggtign. Likewise, when an interviewer reported that the reSpondent either refused completely or greatly Opposed his attempts to conduct the interview, this difficulty was referred to as a ppoblem of opposition. It was found that 1h% of the interviewers reported problems of communi- cation, while l6% reported problems of Opposition. Three major limitations to the data concerning reported interviewer problems were recognized. These all had to do with the fact that they were reported rather than observed problems. The three limitations are: 1) the interviewers were volunteer, previously untrained interviewers who might not be able to recognize interviewer problems; 2) in some cases, 108 the interviewers were not contacted for more than a month after the interviewing job was completed and therefore, might have forgotten the problems which they encountered; 3) the group that was trained by the Michigan State College personnel were warned in advance that they might encounter these difficulties, and therefore, they might not have reported a problem unless it was more serious than they had been led to expect. The Chi-Square statistical test was applied to distributions of the types of interviewer problems hy each of the variables to which they were expected to be associated. The Chi-Square test was selected in preference to other tests for association, because it is particularly adapted for use with non-quantitative data. The negative findings of this study are almost as significant as the positive findings. It was found that none of the social character- istics of the interviewers were related to interviewer problems, which means, that a broad base of volunteers with a wide range of ages, sex, education, and location of residence can be encouraged to participate in a self-survey without fear that these factors will operate to their disadvantage in terms of interviewer problems. It was also found that the previous experiences of the interviewers in community activities, local church or school surveys, local organiza- tional projects, occupation, etc. did not contribute substantially to the interviewer's ability to overcome interviewer problems. Therefore, it would seem that peOple who have never taken part in community affairs could be encouraged to take part in a self-survey with little chance that they would find it any-more problem-ridden than the experienced community participants. 109 Finally, it was found that there were no significant differences between the following groupings in the extent to which they encountered interviewer problems: interviewers and organizers; interviewers who completed five questionnaire and those who completed twenty-five; inter- viewers who participated in surveys where involvement was along friend- ship lines and interviewers who participated in surveys where involvement was along association lines. However, it was found that interviewers who were trained in any way at all for the interviewing job reported problems significantly less frequently than did the interviewers who were not trained. - 'When the factor of training was considered further, it was observed that those interviewers who attended training meetings reported signifi- cantly fewer problems than those who were trained individually. It was not the fact that some of these meetings were 1ed.by Michigan State College personnel that accounted for the difference between those who were group-trained and those who were individually trained, for no sig- nificant difference was found.between those interviewers who had attended one type of meeting and those who attended the other type. Rather, there were aSpects of the training program which seem to have been more available to the interviewers who attended the meetings than to those who did not. These werei having a practice interview and reading the mimeographed instruction sheet. Both of these experiences were found to have a significant association with training problems, and both were found to occur more frequently when the interviewer had attended a training meeting than when he had not. There are probably many other 110 things that were covered in the training meetings and frequently omitted in the individual training which could not be analyzed in this problem. Many of these might also have been related significantly to the factor of interviewer problems. The two essential implications of these findings for self-surveys are: 1) that interviewer training in regular training meetings should be encouraged, for interviewers so trained report problems of communi- cation and opposition.1ess frequently than do those who were individually trained, and 2) that a survey expert can probably train a group of the organizers to in turn train the remainder of the interviewers and find that in terms of reported interviewer problems, the effect of the train- ing would be the same. Since the principle of the self-survey method demands that the outside expert's participation in the survey be kept to a minimum, this is an important finding. The implication of this study for professional agencies concerned with the interviewing process is that interviewer problems have proved themselves to be a fruitful object for scientific research. It is sug- gested that interviewer problems might be useful to agencies in.attempt- ing to measure interviewer success. If interviewing quality and skill is related to interviewer problems as Knower suggests,l then knowledge of the factors related to interviewer problems may suggest methods of training interviewers which would have the ultimate effect of improving the quality of professional interviewing. 1. Franklin H. Knower. “An Inventory of Public Opinion.Pollers' Inter- viewing Problems", International Journal of Opinion and.Attitude Research, 5(1951), p. 221. BIBLIOGRAPHY Bathhurst, Effie. Petersburnguilds a Health Program, Federal Security .Agency, Bull. No. 9., 19h9, United States Government Printing Office, washington, D. C. Bennett, Archibald S. Report on Researching Researchers, (New York: Bennett.Associates, l9h8). BOgardus, Emory S. The New Social Research, (Los Angeles: Jesse Miller, 1928). Cantril, Hadley. Guagipg Public Opinion, (Princeton: Princeton Uni- versity Press, 19DDI: Cherne, Leo. To Fulfill Our Promises-eAn Examination of CommunityiAudits, National Citizens Council on.Civil Rights, New York. Davids, Richard. "Your Answer to Health", Farm Journal, November, 1951. Fisher, R" A. and F. Yates. Statistical Tables for Biological, Agricul- tural,_and.Medica1 Research, (London: Oliver and Boyd, 1938): Table IV. Fry, Luther. The Techniques of Social Investigation? (New York: Harper and Brothers, 193H). Guest, Lester. "A Study of Interviewer Competence", International Journal of Opinion and.Attitude Research, 1(19h7). Hagood, Margaret. Statistics for SociolOgists? (New‘York: Henry Holt and Co., l9hl). Harris, Natalie and Gordon Connolly; "Suggested Symposium on Interviewing Pfobigps", International Journal of Opinion and Attitude Research, 3 l9 ._ Knower, Franklin H. "An.Inventory of Public Opinion Poller's Inter- viewing Problems", International Journal of Opinion and.Attitude Research, 3(19h9). Larkin, K. 0. £9; Volunteers Who Interview, Volunteer Bureau welfare Council of‘Metropolitan.Chicago, Chicago, Illinois, l9h9. Lewis, Donald and C. J. Burke. "The Use and.Misuse of the Chi-Square Test", Psychological Bulletin, November, 19h9, Vol. h6. 112 Lundberg, George A. Social Research, (New York: Longmans Green and 00., 19112) . McCormick, Thomas Carson. Elementary Social Statistics, (New York: McGraw Hill, 1910) . Maloney, John. "Interviewing Problems" , Central City Conference on Public Opinion Research, National Opinion Research Center, University of Denver, Denver, Colorado, 19146. Mosteller, Hyman McCarthy, ep__ al. The Pre-EILection Polls of 19148, Social Science Research Council, New York, 19119. Radvanyi, Laszlo. "Interviewing Problems" , Central City Conference on Public Opinion Research, National Opinion Research Center, University of Denver, Denver, Colorado, 19146. Sheatsley, Paul. "Interviewer Characteristics and Performance Part III', International Journal of Opinion and Attitude Research, 5(1951). Witt, Edith. "A Symposium on Interviewing Problems: The San Francisco Pr0gram for Improving Standards", International Journal of Opinions and Attitude Research, 3(19h9). Wolfe, Ann G. Teen-Agers Look at Their Town, Division of Youth Services of the American Jewish Committee, New York. Wormser, Margot Haas and Claire Selltiz . "Community Self-Surveys” , Jahoda, Deutsch, and Cook, Research Methods in Social Relations, (New York: Dryden Press, 1951T, Vol. I1. Wormser, Margot Haas and Stuart Cook, "The Use of the Community Self- Survey in Combatting Discrimination" , paper read at the Fifty-fifth Annual Meeting of the American Psychological Association, September, 1914'? , Detroit, Michigan. Wormser, Margot Haas. Northtown Surveyp of Human Relations, Commission on Community Interrelations of the American Jewish Congress, New York, 19147. I Wormser, Margot Haas, and Claire Selltiz. How to Conduct a Communipy Self-Survey of Civil Rights, (New York: Association Press, 1952). 113 Agency Publications without Authors Douglas County Rural Youth SurveyJ Rural SociolOgy Department, Agricultural Extension Service, University of Wisconsin, 1933. Ross County_Rura1 Ybuth Survey, Ross County Rural Planning Committee, 19111. Let's Look at Ourselves, National Citizens Council on Civil Rights, New'York. Have You Heard What's Cooking?,.American.Jewish.Committee, New York. The Negro‘warJWorker in San.Francisco, A Local Survey, Race Relations Department, Fisk University, Nashville, Tennessee. Pittsburgh Study, Race Relations Department, Fisk University, Nashville, Tennessee. Minneapolis Communipy Self-Survey, Race Relations Department, Fisk University, Nashville, Tennessee. Montclair.Audit, Montclair Intergroup Council, Montclair, New Jersey, 19D9. You and Your Neighbor, Agricultural Extension Service, Ohio State University, Bull. 307, April, l9h9, Columbus, Ohio. Clinton County Health Survey, Clinton.County Health Council, Wilmington, Ohio , 1930 . Appendix A The Lenawee County Health and Defense Survey Sponsored by The Lenawee County Health Council 1 CIVIL DEFENSE INFORMATION—(For use by the Lenawee County Office of Civil Defense) Were any members of your immediate family in military service in either World War? ( 1 Yes ( 1 No Of those in military service: —were any trained in decontaminating an area that had been poison gassed? If yes, give name —any trained as Medical Corpsman? If yes, give name ----any Laboratory Technicians? If yes. give name —any Army or Navy Cooks? If yes. give name —other Speciality? If yes. give name of speciality and name of person with speciality Efforts are being made to inform people by way of pamphlets. newspapers and movies about what to do in case of a direct bomb attack. Have you obtained any of this information about what to do in case of CI bomb attack? ( 1 Yes ( 1 No In a war emergency. how many extra persons could you house temporarily? Would you prefer ( ) Men ( ) Women ( ) Children ( 1 No preference If answer to any of the questions below is "yes" enter the name of the family member at the right of the question. Are there any family members who had training or experience in Red Cross Canteen Service or Group Feeding? Anyone completed a Red Cross First Aid Course recenty? Anyone interested in taking a Red Cross First Aid Course? ‘ Anyone Completed c: Red Cross Home Nursing Course recently? Anyone interested in taking a Red Cross Home Nursing Course? Anyone in family qualified as a practical nurse? Anyone in family a graduate registered nurse? ................ Anyone trained as c: Red Cross Volunteer Nurses Aide? Would this volunteer Nurses Aide be available in on emergency? ( )Yes ( )No Anyone in the family trained as a dietitian? Anyone in the family trained as a home economist? 1 Anyone experienced as a hospital laboratory technician? We would like to have your name and address. It will be used only for the purpose of reporting the survey results to you and for Civil Defense purposes. This sheet will be separated from the rest of the questionnaire. Full N ome Telephone Number Address in full School District Township Sept. 1951 The Lenawee County Health and Defense Survey Sponsored by The Lenawee County Health Council Township Name of Interviewer School District Date of Interview Check whether home is in (l) ( ) Village. town or city: or (21 ( 1 Open country. Occupation-Head of house Fanny (II-15> ‘3“;1" .24.. if... :2: .63.: Numberinfarnilyathome ( 1 ( 1 How many in family now in school ( 1 ( 1 How many in family had general physical examination since last Christmas ( 1 ( 1 Immunisation Status . (ZS-301 ‘93.?” .21.. £23. 33231 .63.? Ever vaccinated for small pox ( 1 ( 1 ( 1 ( 1 ( 1 —vaccinatedinpast5years( 1( 1‘ 1( 1( 1 Had Diphtheria immunization ( 1 ( 1 ( 1 ( 1 ( 1 Had Whooping cough immunization ( 1 ( 1 ( 1 ( 1 ( 1 (46-501 Had Lock jaw (Tetanus) immunization ( 1 ( 1 ( 1 ( 1 ( 1 Past Illnesses Has anyone had: 111%.: rte‘ru yialrz 3334: 063: Tuberculosis ( 1 ( 1 ( 1 ( 1 ( 1 Hadchestx-rayedinpastSyears( 1( 1‘ 1( 1( 1 Diabetes ( 1 ( 1 ( 1 ( 1 ( 1 Cancer ( 1 ( 1 ( 1 ( 1 ( 1 (71-751 Undulant Fever ( 1 ( 1 ( 1 ( 1 ( 1 (16-601 . . Polio ( 1 ( 1 ( 1 ( 1 ( 1 Typhoid Fever ( 1 ( 1 ( 1 ( 1 ( 1 Rheumatic Fever ( 1 ( 1 ( 1 ( 1 ( 1 AnywhothinktheyhaveHeartDisease? ( 1 ( 1 l 1 ( 1 ( 1 . Anywhoknowtheirbloodtype?( 1( 1( 1( 1( 1 Medical Care How many times have members of the family seen a doctor or practitioner of any kind. either at home or in the office, since last Christmas? (11:13: 11:11 Vault: 323: 063125: (41-50) Number of office calls ( 1 ( 1 ( 1 ( 1 ( 1 (SI-601 Number of home calls ( 1 ( 1 ( 1 ( 1 ( 1 Is the doctor or practitioner your family usually visits a medical doctor. an osteopath. a chiropractor. or some other kind of doctor? (11 ( 1 Medical doctor (41 ( 1Naturopath (21 ( 1 Osteopath (51 ( 1 Christian Science Practitioner (31 ( 1 Chiropractor (61 ( 1 Other The following group of questions apply only to family health since last Christmas Under 1-4 5-19 20-64 65 8:. 1 yr. years years years over —How many saw a dentist for work other than pulling a tooth? ( , —How many had a dental examination? ( 1 1 How many had all of their dental work completed? ( 1 —How many did not go because they actually could not get a dental appointment? ( 1 ( 1 ( 1 ( 1 ( 1 —How many were admitted as patients to a hospital? AAA V 1 1 UVV A AAA (Do not include T.B.. Mental or Convalescent Care) ( 1 ( 1 ( 1 ( 1 t 1 Cause: (1) ( 1Pregnancy: (21 ( 1lllness, accident. etc. —Total days in hospital: ——l.n Lenawee County? ( 1 ( 1 ( 1 ( 1 ( 1 —OutsideLenaweeCounty?( 1( 1( 1( 1( 1 —Have you or any in family been advised to go to a hospital but did not go? (11 1 Yes (21 1 No If advised to go. but did not. why not? ....................... (31 1 Finances (41 1 No beds open (51 1 Refused to go Sept. 1951 (61 1 Other AAAAAA Some people carry hospital insurance; some do not. Do you or any of your family have insurance to pay for hospital care? (11 ( 1 Yes (21 ( 1 No If ye. ask: Under 1-4 5'19 ”-64 63 & 1 yr. years years years over (49-53) —Which members are covered? ( 1 ( 1 ( 1 ( 1 ( 1 ‘ —How about medical and surgical insurance? —Have medical only? ( 1 ( 1 ( —Have surgical only? ( 1 ( 1 ( —Both medical and surgical? ( 1 ( 1 ( (691 Would you say you are familiar with the services given by the Lenawee County Health Deparunent (l1 ( )Yes (21 ( )No If yes. what services have you used? 1 ( 1 1 ( 1 1 ( 1 ? VVV MILK SUPPLY Are the milk and milk products used in your home (11 ( 1 Raw raw or pasteurized? (21 ( 1 Home pasteurized (31 ( 1 Dairy pasteurized If Farm. Ask: How many milk cows do you keep? ................ In the past year have your cows been TB tested? (11 ( 1 Yes (21 ( 1 No (If yes ask:1 How many were lost through this test? ............. When was your herd last tested for Bang's Disease? .................. Any of your herd certified as free of Bang's Disease? (11 ( 1 Some (21 ( 1 All (31 ( 1 None Mastitis is quite a problem on most farms. Have you had any trouble with Mastitis in your herd during the past year? (11 ( 1 Yes (21 ( 1 No Have any of your dairy cattle aborted during the the past year? (1) ( 1 Yes (2) ( 1 No (791 Have you tried any of the vaccines for Bang's Disease? (11 ( 1 Yes (21 ( 1 No SEWAGE DISPOSAL (161 What kind of sewage disposal do you have? ............ (1) ( 1 Privy (2) ( 1 Septic Tank (31 ( 1 City Sewer (If septic tank. ask:1 How is the overflow disposed of? (11 ( 1 Open ditch (21 ( 1 County Drain (31 ( 1 Underground WATER SUPPLY What is the source of your water supply? (1) ( 1 City (21 ( 1 Semi-public (Group) (31 ( 1 Private Well (If private well. ask:1 What type of well is it? (11 ( 1 Dug (21 ( 1 Driven (3) ( 1 Drilled (41 ( 1 Spring (If source of water is other than city, ask:1 Has your water been tested during the past year? (11 ( 1 Yes (21 ( 1 No (If no. ask:1 Do you know how to get your water tested?.... (11 ( 1 Yes (21 ( 1 NO DO You have running water in your home? ........................ (1) ( 1 Yes (21 ( 1 No If you live in a city and the city is not supplying your ' water. does the water main run past your home? ............ (l1 ( 1 Yes (21 ( 1 No (31 ( 1 Dont know HOUSING Do you own or rent your home? (11 ( 1 Own (2) ( 1 Rent Do you have in your home or apartment: —a flush toilet? (I) —a wash bowl with running water? (1) —a bathtub or shower? (1) —a kitchen sink with running water? (11 1 Yes: (21 ( 1 No Do you share the bath with other families? ........................ (l1 1 Yes (21 ( 1 No (If yes. ask:1 How many peOple altogether use the bath? (311 How long has the family lived in Lenawee County? (11 ( 1 This year and expect to move (21 ( 1 This year and expect to remain (31 ( 1 One year but less than 5 years (41 ( 1 5 years or more (If family has been in Lenawee County less than 5 years ask:1 Where did you come from? (11 ( 1 Elsewhere in Michigan (21 ( 1 A state adjoining Michigan (31 ( 1A state not adjoining Michigan (41 ( 1 A foreign country 1 Yes; (21 ( 1 No 1 Yes; (21 ( 1 No 1 Yes; (21 ( 1 No AAAAA '4 A) endix B .5: *‘d ..l U- ‘ ' _ . .. o" 1 1 A. -. 1 a . . ' -- ,‘ I . . Q I L .- 1 t' F‘ ' _-\ . 3 "’ r. ( wk. -1 \. J4: .. ...~ 1.. .... . ' - .' ”Vtuowvwnss' ~fli\’~0‘¢“- 'I'Miw "\ . (a n.‘ ‘a'I-‘-" r- vfl u» -u¥ . .9 - .gn ‘ .'s‘ 1.." ...‘r 11 z, .— “4 — 1.; . . . - n . .‘v-‘ufi..vv-~.v s. .ru"O§-‘O—...-e~‘ --. v 3‘; 1 v'T‘. y ‘ “ 'fl;?”l ~7r’;l' : sir-ff“. ‘1 ;.\~ .“I\“ry‘\‘ ‘_\ '3‘]. ‘1 "":.“(’to \ . . ‘ . _ _ ‘ ' . --tui .'_ nun...- L- .1 k1 _ - 01.-.: "“k' J-vls .1.n..'..' 1.1 ....-..- . ‘. ‘1 . . . - (I \ 1 1.2!” .4. H... \V Dina-h H-.—le---o..4.~. -... - - “— own—44....- .4 . ‘ -., ‘1’} “...": 11mm '- ‘u ' v- r‘ e '. l .. 2 I ~ ‘5‘ r ‘3‘ r".“t‘ln ‘4 Q u.‘ ‘).f‘ I‘ ‘ In. .'\ . - 13"! all ‘-“.}L‘v,’ I'A" ‘ ,. .\.Is .1.!’e " I .‘tfl..-”0~~u‘-'- -.-..ou-p aura-pvt. Os-gvv-u- cat-..-. 1.. l‘ s o t . I ~ - .fi-n \‘ ~.- .7 ~«.-4,.1 r.*_~,,~n ftpyff‘uyv V.) 1- 1.1.; ’ ,. l,'-,'.fi '3 .\ 1' .‘A'..el.~.. gut. I s!“ , | . on n..*~.e.-..--..~-«~-- I 'P" ,1. . a..wr'1‘r~-~ ' .. " l~~ V v - ~ . . 4 . .~ I .-\J. ..J- L'KIJ- ‘a.» ,,.,‘\ ‘. -‘...' k’. ...-4 ‘ s»..—-'--a..\ “—bmfiw. -.oiuaum.-. **-fl-Lu~ ca..- » t I .-.AJ«‘ tyre».- .. )_. , 1) 0"- -~~v“l m ‘ --.‘hr’ d#‘~.i<1"‘d.°.-—~”' A “- N " " t ' a 5. .‘ -3 .l‘hr. ‘,“. .. \J I. s [.V ,1 .J‘.\ a a ‘9 1.‘ it‘ll- 3 M'H-QI‘.’ «01- A; g “3.. .. 1 .r‘ s ‘4! .1 \J -AIV" “ "IQ-...f’flfl -«fivh ._._~ -.... , . 2., fl 1L1V L,OJ ... ' ‘ .e-J T I . a. ‘J‘ic 1 ~31 '11:: .u‘r.r- .0» "-I- 0 - ‘ .-. 1-1! ”f. ‘hflgzt 14". '-J'1t':.u|.o ."~O ‘- LAM- liz“ ‘ ‘ 'I I_‘ . y _ toe naensn survey Lu: ‘ . I I’ 1 . v- t :9. 1331313531" 194‘... or "Hm-1‘. ‘ v 5 \ ot.=1e r ccmatz. res tn; ‘ - tire»:- vre «Wt 11’. 711.": 4' 'fy Tr, T" -\“ r'.‘ v17 .91; M “A (111.30.; L .Litlllu‘s ...1-'-".-:"i-1 .L.‘. 1' , (1).". ”:A'Orru rips . ounce-r 4“ ’7 tr 1".) “an: o-zoq-w')‘ .h‘4~ 'l.‘. m. '1’ ..,_ . ~ q. .. .... I V“ alt ‘I‘ I | I i ‘ | kfi'kt. . 1'4 " -~. .l 0. vi:- LJ out-.-.n—w-nu-e u “WW: -¢'um--- '-§-0'~‘-' A rvhdv know. D ;_‘,. v’. 1“. 3 yolk gr: \I ;:u t.: the region :nu use prlA " ' ’1 "'5‘ o .o-‘ trite)“: 13.1-71 (5.) ask .19 {tcfree Jr to itmn 9 l‘ I 3' .7 D. Ah-v‘i... :‘ ELU‘J p .‘.;t: (flu-1r i“ .UU viillis11Ci‘: 1.61Jn it we mav be able ' l \ .- 4 uv" :tr')L‘t7 'x“?.,. ”the bot-321.81.. litifiC’iZ'fll L>:%'v."'r_7_:3t33 1;. we are very much inicrgstsd 1w We would 111:0 to he": to give 5' 'sstiens '. r237: b tit 53113113330 1:. doing: the same thing... .3 ask you sehetb-nv about;unn'csn.i work on the surre Kunizer cr‘e°t ?0n, vx.‘ 2 3} and nut r ' Cthw 1'1‘.':i.:tt3 vl '5.ly an or £1 sing F l 1 l - r) 5-4 I O ‘ 1' j; .(v .. -;L (2)731! ' . n '2‘ --r.~f- ..‘LJ ”,3. .‘,.\' "L ” ’ ' “’ “0 ‘ " L ‘- ”Ct-VJ 1‘ .1 “'I'JLHU.-1.19 (“'1‘") I I I t u '~ ~ .‘ ' '-' ‘. "! . n " “ -.|.. 'p p r v ; . r- .l - ,4. 1 .. A. I 2 -. \b s ’. ‘. I. .1113 1'.‘ . ‘, L 1" y I ,1,) Aju,’ Jig" I‘ 0:: .‘1 111’.H:.‘,¢_'~1 .L .L tl’lk'" . . . - ‘ .. . l _ J _" f‘ L w -'\ 1 . e . . . t‘ ". 1 t :3 ‘ .. “ f“ ‘ ~r 3': .e- ' - \. "-u --,.—. . '9 , ,9 I-, ‘- . ... .. v .‘fi ‘... L. . .1_'_ -. ..3 2 .10 . ':-. r... ‘ t’ 721?} VJ. 4&3. _ . _'_ 1’ ‘1 I" .~ 1 ~ _r so. ’I ‘ ..l‘ . , 1.: 1”“ )4 y A. " I ' V: _\ _ 1 j 1“: 11:11.9: ~,.~. 1' 1 ~— ‘5 4 'J -- i. v r h! ‘ Q-v :y-n - -111» e a .A4A|1-.. o»-~§-o. . -. .; 'p-e- .2 , 9 v. , I .0. 11“! ‘l *u» . . .1 L -‘ . . , 1‘ -' * :".“ ' *I ,' :131 \ " ' ’I " . ' ' - ‘ '1 . 1l.'.‘ I“- L ’r‘ J u‘ , U..‘ '.«LA I ..‘ 1 Ir .. fl . ".1 .J.11. V , “' I_‘ v - Ju. - (“'4' “‘15.; u..¢.wn ...,nnx.» act-m" .-u-a-nmm ( . . . .. . 1 {q . 141,110 .3». n‘ I ‘ ‘3 I ‘4 mm ..--vro. . ficuna ‘..*"’-lm‘~-fll ca ... . ..'. ‘ -.. fl‘ 1\’\ ‘ I"‘ N LIMA}; u“. ‘4 'H... - 1) n 0“]...- o.~.~.. I-OA 'qflfirbL «ma-m. ' '~‘.‘I\' .“ ‘1'", w-V‘L'I ~‘ I,*"1.‘.' 1". 13“.“.[52 . "1‘53 ,._'1._'. :J t 4.,v‘., FT. ‘0 ‘0- OLO' ' ' N'IH'\O'-‘C.'.hb¢. I.» .bl-"‘-""‘l '. " ' ‘-\';/- :Q -‘.w .1." non-'4 -ro-Os-r‘u -....,_.,- s:.p«-~-..n‘--~._y-.--s.‘ al‘ v 1"“) val“ out. . ro—--‘-— ”4‘. w - u - i‘Jo~..n\.ro»U--Q- ; /‘ " r 3 - - - ‘I. s ‘I ,.‘ § 7 l I ..-1_.,~ CI" p > p. . 0..) fiw‘ur 1.0. I Is ’t""" “" ’ .. 1" 3T ‘ I ‘4: '2 ‘In‘ I 1 \ \J‘ I .l‘ta. t’ u‘ \21 -;‘l§'~.. c— "1 ' ).‘I -‘ t,‘ ‘- -. ‘ ‘ ’i i h' l ‘ K'J l' . 5 - .'1A ' ' ‘ ‘ ' ....A..~, I‘ C no. ‘»- an -- --.- -1. -.-....- . z... - .J... . . ’7 :1 ‘ v" .. a 'v -.. f n', . v, -.- ,.!, L *1) «-1. Ii ‘. J '1 . .L. "‘11) “(1 r' ....q. {1. ‘ . 1; .3. .‘_ V '. AL; ”1.": .1 ...111‘ J '1 .'.v.. 1 J .1 -, N“ l ‘ . '11 1' ._,J \1 .04» --\r' ‘ ~¢-..Is-.( n .~ ' 'I ' "- .".( . L.) .f I". I“ .\x I wrv. ”"0! p ’2: ‘ ’r_ ;_‘1‘ H .f‘..“ t‘ ',‘ ,1 ... o~v. \ ‘ ’ an... a". J ' ll '1 o- ' V \ V ‘ I P. r \A‘ I . I: u, . . . . -- u -. - .. -< c» w . 1o- '1- ,. ‘- ‘ ‘ ‘ .' “" “y" . ’ ”‘ (2‘ 1. 1' "*i i. 113' 1V1 . i ""’ LU Mid. 112.”) ~ L'L'] a). 9.: " MAJ. ») f‘ I.. GO 1;” _.. . l! .. . -. 1 L .- . . ‘ «. f L .‘ ~-v.1’ — . ,t'gO I {1" ~. - . l \o‘ .I - ... ~' 3.» a ‘1 "1“ I. 1" I \" v‘ :1. . I Lr~aug ..x1. 3 I I . 1 . .: .1 r‘” -. -1 . _ . .1. .. .. J. - x .L} ... ‘ I‘Y. .. I:.’“‘ “"I 1'? (3" w‘ 1V 1 s. . an \m; A W.:1 :31lar . 1113 m,p0¢ngb £01 U131 a“ . . . ‘;_‘ \ v-~1 ‘11—1r F) -» NV I . . v ‘ ' .J 9 "' a : ~ ,‘ ' " 1".1 1.3.1......1-~.1 b 1. 9173111112 I .. ‘l ‘ ' W‘- .1 a 1 ‘~ \ a a a '- u-ua.‘ I r In- . . .1. ‘ '3’ ‘ ‘ 1 _ I ‘1 I P a , b K :1 '1' O .. L¢-LO‘ - ‘1 ‘) ._1L,‘A’N lo 0-). —‘ ‘ic'ld 1 \o |'J\J. O'aé ‘nle I.) B r I \ 1. l I L ' - ... ‘ v'. -: - -o .. ‘ ‘L 'n-i 4' . -n 1 ‘ upw- P_~10f“‘1‘ 14."; ..‘._'.x. :.".. J,‘ .....(. La 13:11-1. ".3, :u.,.. :n 1 I 1 T 1 f1 .1_‘. _‘ .. ‘..’ 1";/"‘"1}"‘ 1 V?"} ‘ 'C . f“‘1 ( . 1‘11. r ... _' “ ‘. _I‘ "5'“! + ~ .I ha -- . - . . ' 1- . . I' - . ‘. ' . ' .' . - \r/ ‘1‘ Jl-'-| :- ',‘.J\‘.( v]- t" '1. .lg‘c’ul ‘1- ... . ‘4 J V g. 8010 film‘s—V 21"... (I: C0 \ lid-um ’1‘0- o—u ‘ 'u i ‘J U .$\l J- n: I {'3 “" F u 1‘ -'Dw::v( :(3 1‘ '3 ‘0‘ 7.; ,t‘f’Il‘ H 1 (.+’q0 U J... 15‘.) t A...» U l .‘a‘~‘z‘. ....1‘..!.'l' Jaul Al). (I I 'l c:" u .‘ r In. ' f) p . ‘ ‘ .- - ‘J - .N c _,. - .5; ‘ ‘J .' -'- -- l. . .1: \.l ‘.' .‘A 'I {1 ii ’4‘:- r ,. ' . . 1 . ' i .1 i ‘ . ..‘ ~ .3 .31.? = ,i I. r, .. 1:, 1~-1.,.A»u-=.u ' I o I-‘r.'~v~ fink, , I: I :3.l‘; \l» s’} . o u 1"] . : L. 1"”: g .I- ~‘--. I y o 11 13 3.3 .112) f n 7 , 1 r - 1 o 1 r. r, - a A . u - n I V I. 9 '1 . ‘ . (.1 '1.-‘,IJ. ' . '5 : Q'J" 71 J ,- .. .‘». «.16. -\-.’,-‘.7 .‘ ‘v’t' -.‘ 1'“ ~_ , .‘.-.I ._ EA: ., ~A an“ wu~~ w1~v wu- LUCJIAUA 3AA1“9133L 11%1HALg 1L1 an: 14.31:,1<:{ J u C ' ‘ I I \ v Q « If: a ‘ ‘i""“- 7‘”; I ,3... .‘. .‘.‘ ,fi (13.1,! ,i) ii "’12) “‘1 . "".3 )1.“ .i.‘ A 3&1‘311‘. 71‘ a Y"? I.’ .71'.\'71, I ‘7.~‘a 1 0,). gr, (- fl ”.4. ...-.4. ‘71.. \‘. ...») b.» .".'.- \‘./ I \U ..L.\\ L’. '1‘ J. '.-. L‘..o., L04 \4..\§ L1“... L1 1 ' 4 :. .‘ ..- .s 0. " 4" '2‘ .7 up:- u: 3‘ “'1‘ 1.1“": _‘ c ,.i 3' '0I .- .- tws'l fi..; :3"... I '0‘..1<: ."-l:‘.'>" E46 IL U‘: II“; o‘-_ ."‘. ‘ \ . “ ,— .W ‘ t ' :-~ Ho. 8. .J.J -t..'. -'.t‘.'1» ‘J- .Ll. { I” v..‘ \)~4.. nh‘v" ’1“ ’¢ 1 ...-’A :5- ‘\.—’ - I ...1\ '1' t ’I I I' . _ -. n .1 p ~4 ' ... A. a I v v v 1 I 1. f I .-. l-_ - . ... "T ‘ . r._.... 1 IQ. . 3 .. w. . .I -g a. .- 4‘. — I- , - , 1 ., . . ... . - ' - . u f' -. 1 , -. V. _. ,.q \ r1 ‘ (3.0, Q. “ r 9 s r SI|r\"1‘r~ ‘i J ' ' '1 -'--"- ’ ‘Ja "A VJ t'r' l)-‘o TE CQI:. '- ..ac.‘ \Is.4 u-.."- '-.v‘ -‘.'I.' ("'J‘.) ..t.' UK. an"? J...L (A 'LL‘V " ‘3 ' ’1.” ‘ ‘ _’ ' ‘1 -\ I - . I ‘ ‘) a ‘ I \ . . "I ‘ ’ I ‘v 1’ Tr i" "‘ "'3' "'.\“1: ‘L‘ PINK: ”1 '1 " ‘5' ’."' "‘H" ,‘_' ’\ I ' V1 1' ‘ “ '1‘ W‘l '~. ’7‘ 1" - " "‘ "" ‘ ‘ " ‘ 1“ ‘ vu’ ‘ t‘ .‘ k' ' 1' -:\-‘v‘\ .A ' j. ' O . '1‘ .f 1.2 .3-.. :-u '- ; J. ‘8': :1 .J'.{..' (Id 'c "1‘ " no... ' \‘ 1‘. t.- l~.,l} ‘1 3133' 3' I4 ‘ I a .L', .l ' '1 1"" ‘ ‘ ' . -4 " A ‘ a. \ u a V . . A .. q , .. ' ‘ ..‘,.\’ . A. . u’ .03, . ... . ,r . _ '. ' ' ,- J - .i .l.('. L .'. 1.. . k1 . .u 1.13.115". A... .~..‘\) 1.111 .1 1).. .5..- - 1 15. 160 173 . ’ ' . . f‘ . '2. ~ . - 1 ... “‘ H . a‘. " .. ' a .‘ ._ '4. .fi't2 1H~firx r ( n? ;uu'z‘) a ('0‘9! Oi LLu uni? a unu qiLaeienAeixc; v o 2 g s ‘p .0 :3 -.\‘o I .2, _. ‘ ... ‘~‘ . H . 1‘ 0'2. 22.-213.022. 21:26:15.2;723 ' you firm it 22.3722 72.7. 2.227222. - Lott». 23.22..~!Lf'3! 1‘ ." " ’ p;. ' v ". " ~ :‘ '2‘ ..‘2 ‘ " '7‘ (2’ .e 27026:. 1‘ 2,177. o: a monk cor-r of 2:23 2 e2. 2072.22.31.26 1. 1.2.7.22: me \ criticimze12.9.27 at 1;; 2.027 to this bonnie“ 0 .) (1.) Why did people find these qneetioes 22.52.17? 10.23.22? (2) Weie there any questions 272': 2.022 you did not 2277 zerstartd? A U) ~_.I ADJ I'Ihich you 0 222.72 t 22.21777'12md 721;; it "are asked? «...! Pr'o’zze £7.32: a- profile-.225 07'. 2.2671025? 23, 13727221227373 0: blocks-7'59 0. pr 0 1 '2 :23 vi axe-.5100 ' ‘ ' \ ' V '2' ’d 1" '- ' 2" 4": ‘ '2 ‘ ° ’- 12... 23.2.'o:..2.2.‘~.".s OI. 012.22.2221203...27...; ....2e neeemg 2" r... :9... wheat 2.2.7323 etions do you have about ”new the one: tionne .ire 12.2.2; ht have be0 i'nm'omd? (Iota to interviewer: Again make notations on plank 0011‘ 91.20;; questi onuxire) In general, how do you think peeple felt about the idea. of a health survey? (Note to interviewer: Again keep in mind criticism of Specific questions which were cone idered silly, useless 3, nosey, Record these on questiormairen) Probes Did peeple feel that the questionnaire included the really important heal 7.21 questi one ’2“ Of the famiies you eporoeeheo‘. 2.70:7 many refused to be in. ‘ 12:71 ewed? ‘Cfnet were the 2.208%. c .2772 3:. reasons. for 1"2..f.'1..-.i}“.fj? "m “0. About how 2.20.22; ' of .232. people you ifiterviezwed 17:273. heme abe'ut t..0 survev Infore you arri'eed {227' the interview" “139 w:—“‘ —- ~— .2 ‘ ' | v ' .»—~ 3; -y 3 f A > I x a g I l’ "l I I ‘ ,1 1 I ’ \ ... I" H15 1 ‘wk 1,” ‘ '1‘ k ‘ “I ‘ I} l. \ L x , I . A .'\ 1 I I J .“ , . J .' . =~ v . 1': 1. y l a v L‘.‘ J .I ~ mu 1 ' 1, ).i. A] u a ln-isa» \‘ \ .30-1.1: .L - v -~_a—-.--a:~ '01- .. Cid.‘~l-‘- -y m ” I“. . \n- .... a‘“. lull-.... uswom-A ( I. ' _ -.....m..- (3 i , I 11;! 4' EL, .) :L;\ 1' l. w .- -— .fifi“~dr‘-.~-*l‘i’--Inw¢ 'uld you '53 .1 me 11 212 you h."- '* o 1L1331L (Start from the thaw ;'ou first 3 ed to bee n, aciive inthe SHIVQS? 8rd abzmh it and. (Is-:51; its each step. ) ‘h ‘v I 13019 to '3‘31‘V.‘.?J“. 32': Be H9 :d vrr )flu 1m res pm :39 Pram f0? : lo Villen did he first. hear abaut the survtp idea? 91 h?” “id be Happen to tage cart in the :urVOJ? I5at3I fer: a: I.-,-~_ awash; ‘S 'to individ 1338 (0114: akin namrs) bn Juufio“ hips no 0““an antic.s (SpeciIE) pa 10131.10nships to feelings of assisting with apncral mxmu.uity ba1tenren no activitiesc Robe to interviewer: When this is fianhed go back over what has been said and reconstrtct each step as a check on tfze accura cy of your interpre1ationo 213 Are share any n1ya1*3aalon have whi ch were opposed to the survey? yes no Don t kno - n11 yes, ind1.caoe be low“ Reason reSpondent Name of Pro you heapwndJnfi's Haw long Reason far participated in survrv organization a memcez? pcsiticn 1n a.member npnccitlen bv, despite the ‘1 q-y 5 l -.- ' H J 'v \ V 2‘, f‘ '0' M'.‘ I‘. F? I‘ .2 I" c {.‘1 {1:141 -&0 I. 1 ~. (h . . r l' 4 1 t .1 '3. Y I ‘ 1 . 1( ‘ . , _‘ j ‘) i . k ‘1. - I , ‘ l . 'J J ‘ - r. .1— . . 1" k. I .1; . a- . II‘ .. I1 _,, . ,v u . .|I,. ., ...‘ 1. “ ,- ‘ r “-, ... ' a . ...‘ ‘ ..- ‘ ‘n. ’ . I '1. -. . --n . .'.. '.~-.*_ . uf. .\ l I ‘J '.. -r I. . . ‘ - , . ~ - .‘1- . A.‘ ~. 3" 1')... £13.,“ ',.,l.‘ 1.- .‘.. fl“... '7. .‘.. "‘ A . I .-J- I I‘. h .|. L-‘ I. ".v. I”. ‘-- -.A AI‘-~! l l-.J y‘ ~ru a... V .. -1 ‘utnux ..n. . 00v 0—. J ~...~.I‘ f~¢v~ ...o. - *to~tl no -w. .4 .‘l‘lfl'J-V .Ih-H' who r.’ .50." 'P ‘No . n' ‘ i \ v 0 e ‘ I‘V‘o. a.» ., 0.1.. IL-yo'u-lms Iouuwm-x-ws“ w {w .w- “,~'m—I‘ ”HQ‘D -.. arc-011.4 w .4- -.lv--v-nnon,—- ...-g. 1».- -‘-9. I- u - I- \- I I i ‘ I t ’ . g .t‘. ., r. own-— II- '0' g'u. upg- ruq- no. ~ 1 . ..- .... ao- 15.». rnr~ ...”. c. s... 'L'1-IV I-vw-r—u .‘ua .g .‘x. .u- 1c.. .n—g - u.- I . ! .b‘c‘r‘cn'uv-r-II-Ml- vu—U'IR II a - ...w‘uQu—m 91-“.- ---- ~wwn- u- a. mhuwn'h‘w D. ,o-ov. lr‘\‘- ‘va-nro "I0 l It‘l‘ oi. --~DI‘. :0 0.. grow u ‘ {v .u»-.n..-I--. - 'ouo—h- 0'. ”n. a- »D---nm I-b!“ gnu. ‘ «_~‘...-_-—. 0.4-. -... I 'un Gnu-UQIM‘ICVQ CD.’ a.- w. -~!Ln u.- v 1'. but! nu.-.~c-v-“~u ”a.-~,. ,» g‘nfle '0 I". 1 ' 'II '4‘ "v ' "I "1 ’E'W“! 1' "‘1' (3" 4". "I'CJ ("1' 3‘ l VII 1' ud J‘ ‘In.'\4.'4 . I: L..~' ° ..n f) w ' '3. XML I B In - 1K L‘J. - I. 1 «K - *2 L L 1| r‘l . _ 4 I J . ‘ 6 fi - 9 .. ‘ -. L . . ‘o v -”v-‘ n I. ,j‘ I! - '\-_.fi ‘n -p I spa l a...) w ., ’n A \. .‘A. \s-;_~_ \ -q 'u.‘ ‘1 Hui]: -.J'.. 1'...” :1 (’1‘: 'J.‘ ‘.. k. .L a). I’;, z Wu} Ls. UL. 3..an 3.. Z. :._‘ I~'1-I~-¢§--"O "‘ .‘. -" P ’ I. -~ ‘0 ' G F-' ' .-..‘ D. ‘3‘.“I3D-cv‘ln- n-Iu -‘ (tn M :4 .L (V . .’\ 5.} -' .' ‘ “-1 " AI, ‘. -_ I‘ (Q .\ r‘ ‘ I ,- ' ‘ 4-._ F.‘ \. .) I \ §~ . \ —_ , o ‘- *3 .‘;-- I ‘ '5 U \ ' t. ‘5 '- ' J. ~at... a 5": ~13. IL‘ I.' '.‘_' ”3“,. \ ,LI s:' ‘JII‘\: \. ; 5‘- .. ., ~ \ S“ .5 -‘ ., . c ,4 ,,.-r .: .... “’7' a} _: 1' “A: 1 a ., )1 - o . . ' - . . ‘ ' .',~ ’ . ‘ x .1. " !-\ '« ..(). 1o'cal\-' \- Lr’--.K0. uJ ,‘. I).L— “J. 'Jo1J-7.) 11.3w ‘1‘)“.m‘n.‘ 'U :-' 1‘ - . 1‘,\ , .: I'i ‘“ 4“ 1‘ n..1 (4 Y , .I‘ 4 - N ' a ‘u ..1 v‘ .V‘ 'L ".- 1 , 'I" .L I qu 1.; .. (=.. .u .‘.)'...L-_\ z)“!.a1".l.~‘; C‘s; ¢Orp13£v O”! . ; cf. .. . ' "2 . J. . ' ' “V .. “ l‘ W '\ 1 5’, - ~ \ - 0~ .. -\ r A-_-‘ r. y .- -- r9. ~10 i - -.a. . ~-- U. .03":- L.‘(‘ . r; L' 35.“ 3 C Ifllx k.‘.'.'.l.‘. Mm. ,_ . 1' J- ..I- , 5 In}; \ ‘I . '1 \. -.’ ‘ .‘ , l. \ I.- ‘7' ,1 7 j" . I‘ I l {I . L‘a-‘a \J L) , Q .‘r. b c: a r x I . . . 1. ' .3 1. .- '1 " .p- .. .1.) - - z .. .‘ -‘ L 1“ ’ V I t a! , f} l ,3‘ I” m I aw ,) i' [C ulx'uwk. p“) F, ...“: L‘. -..a. .z.'.'.'. JR- L (-...J - ';JJ. ll ‘. .A‘ . -. n . K . L 1 ~71 . _. .~ r. a l 3' a --| I» o n I. (w 4 . I" . -Q U. .. - ’ -3 p. . \, J'C- I ‘;{. I! 3. II.)- c L! "..t ' 1.: JV" ‘;Jd!’ _' ’ ',‘ In I I1 ‘ :E' '~ ' : - ' I . .. w-yfi'u‘ n'n ‘ ."“- 'I a -~ “(‘ I r - "" -- -o I ,4 -1 l" \. .Lam. 1' ’..'.J. u (.‘11 i U M“; ..f :. L». ~z-: :: T1? 'C 19.3%: D ., 0:19 , .‘ .3 ‘I‘; N .-I "I’§ ~. A i -' ’ ‘ t In It. - " "1 r‘.'|' “' ’ ‘ " ‘ - y . P. u} I - ,. u ~3 ,- ,vl‘ - .1 ‘,|" ‘ ..u .,, '\ “ _' '2‘} I_,. "A ‘5 ~ ‘ x. " ‘P' i ...r ‘g. I ‘.[A.‘ J t 6 ‘ -‘, ‘ ~ . T. . I ‘.| 3 q . I“ ' . o- ‘. \ - J 5". . ‘ I. I \ , I L4 _ I. , . L , J .. I ’ . o ' l 1‘ ' I ' _ ‘ . A \" r 1 - V K . ‘ K I - I. A1 ' I ‘ n . 3 r ‘ ' ‘ > ‘ ' ~ A A \ \ ‘ I I I A . 1 p I t D‘ I. A Q. I f - fa w‘ l), (w z 4‘. u . ‘ I \,\I . x' , . l '. . 5 I, _. V Cd‘- -.~‘~n'm I- ‘ll -- -. § — 5 ., A ‘I 4, ‘1 1 I ~II'~3 I n’1‘a~‘.’ I‘--~-— O ....l‘.-‘I.I-v.‘ ._--_‘ n'..-v w'oou-u. noIa-\~bu tl-*‘_-‘I r‘.—;-. u-o-o. ...» .... .I. a-’ - — I .nl-rhv-xu‘. nu .'-I~0.'..a .-.4 ...—- ~..~. --'- -o'- ; mv a D r .- 'b'b‘Q-DKQOOI.O~UOO mm.“ kilo. Irv-.... ... .-n - >v- *l-!.! l l I cw“.- fiuvbl.~\I—~ 10"“ nub-u...m~-e—u —»u urn-I: -. a;~~<-fi\' v. I l - __ , " 7 z I I, v} I fl, I. {n “XII: (7.44! fl" I 1:.4. ‘_ I, ' .. I ': r" '5 V ‘3".fi ’vr“ -, ..'..-.".: -:‘. ‘L I .4. , . ' 1'1"" -7 ‘ L L- U .1. s’.‘ . . , r“ ... '1 ‘ r_‘ Cf... 1.4. ~.C-r- . . .. . m n+‘ 4— ~ - ..,_ ..H '5 r ‘3 I.) w; 2111.53 }' (TV/...} I’III A _y - v '3 , h A. - " tf‘ "-n - \ -.-’\ f .1 ,L f}; l-..L1 1.1.- .‘ ,.(z. s ‘ . . . 1 ., -», _—g"-~ 7 . I 7 r- ‘1 I 32),": I If I 1 ;~',.: , o. .‘t“, .I z .. I 0": h“ "' y -‘ Ir 1 ~ .— )7 ' ‘ . I \ n . u r D 9 V - I v 1: 4 u ; '. a H a ... v .. 1 :— Al', ,3 ,“_ >‘ I ' , V I ~ . A . ‘1 \ a ‘ ' ' . I 7 I. ‘ ‘ , .v. . _ ' I . . y ‘ ' V . . . _ . . 5 ~ ‘ H. _ r ‘ ‘ , ' v" ' ‘ ‘ . A. .‘ h. I I a. I ‘ L v~ ‘ ~ ' u. § . . . v. . ,’ I I I ' I > ' K > ' ". v I " I “ -~._. Hi, ‘ Iv . ,, ‘ ‘ . I 3 ‘ r- ‘ . _ ‘ ‘ ~ . .. .. , . ‘o . ,- ,./_“‘ _ . I, - . L ‘ v y 1 . " .‘ l |-- i 4 I .\ ‘ !‘ d ' .fil“h‘ . _ ‘ - ~ . . ,‘ I ’, . -‘. v: ., ... 1‘ ' H ‘ . .v.. . \ A ., 2 v ._ K . ‘ ‘ - u . ' v . . . . ‘ . _ . . ~ a u .' 'I I' ' ' a . . . . . . \~I . \ . I \ ‘ ' '— " ._ . . ‘ P. ‘ —,.~,__. ‘ a V . - .. 7.7 - , ‘ .I! ‘~ v , '~ ., I~ ‘ ‘ .k \ _f 1. . 'l I V ' . V ~ ‘5 u ‘ ‘ C _ . L~¢¢ l u. . .. . I- 4 r‘.‘ 7‘ 4 . ‘o . 5; .- ... .I‘d 4'. w v u \- .. x . . u .3 . ». . u u ‘ A , . . . ‘ t‘ i . . k; ..x ILA v . n I ’ r ' ‘ \' ~ ‘ ‘ I \ I " ' ‘ I I ‘ ‘." ‘ 1 | ' u . ‘ v a. 0 ' I — . - . l- ' n “ Ya . . . _' ' ‘ ,1 Z" " ‘ " . r : *’. '. .9- v-rc‘oo " v‘ . ’ " ' “‘ ‘ ' I I \ g I V \ I I l ‘ ‘. I “ ul‘ l — n‘ l in 0'. k” f (V . ~ ‘ —| "H ' 3' ‘ ’ ‘ ~' ‘1 ,_ ‘... J. . .... 3‘ I '. o ’ [‘3'] 3!. , Q 7 4' 1 I. ; 3 ‘ .‘3.‘ *2‘3""“" "‘1' 1"- ,‘ . 3 ‘l .5; 1‘ ‘ s I, ‘ ‘ ‘ ‘I chi-.0) ~1- r a ,..- 'K .' . \& r . 0-4‘~'— 1’ ...Av..L. -. f‘ «.3 U 1 -. a w: r - - lf‘r :I".I~ ';- u ‘v . ,_ . (‘ -\-. , a s -‘| 3, v c I 3" - " ‘ , ‘ ' .. ' ' ’ A .( x 3-,-- -- . -«-. ~.-- ‘3 ~3‘ ,g. m3¢.-~-- .‘-~~ -“-»1-e~,o»~- 3,“ rash-g! -3 fir: U!‘ 3‘ \‘r’K‘P"V y ~.-.»- 1 ~ '- ‘ ‘2 ; «u L'-‘. " D: 33.1. .-. .T. 731 {M Q J u. u.’~.-.tn~’ .;;.\.r’.l Cm ;,.L.vr; .3 .33}. ’1 I ...}U‘ a \P( - .1“. Lu}? C. .J Lu}. I 3 ‘ , , ‘ -.. . . . . 0 .: , I l O ‘i ‘1 ' ‘ - L‘ an-sw ~ -. I" .4 u . I.» .- -’ r . ., . 3 . .- -- .3... ,u n ... - -o . .— . A “L r‘ "' 3. 13 ‘. .._ ~v . -.‘- u ' ‘ -, . _ : I0 I} 9 , , f. 3, ’, 'r' ‘ V . '1 ' (j‘.' 7 Iv“ 1 v.3. ‘~ ‘I ‘;, ‘5 :L‘L‘ 3'- .‘11 I v. ko-k) O_ -.\.-;.J |. ‘ ‘3" K4 L . .LJv 4.! l, U- L“ ‘10 14‘ '-< J gnu} 1 '(' cor-£- ‘Jll -$—' ~-‘,? -L - "I. ' 1'" n ’ 3.3.1.) . xv I-.. - ‘7 ., 2 fl t 2”,: I. u.- ng ..- , . , g‘ ‘ v ‘- ‘ . . 1 V ‘ s I ' ~'.-'_3 a: .‘ u *3", v1 c; v ’ .0 *‘-q r c? n ' .\ '. 43 ”3‘ r; Ivy \ \fi'-1 1“, vn" ’\ ". "u 7‘,'\“ f,” 2P. , rv 3' - In g'.‘." I)' . .\.1 Y‘.\.' -11.!» 1 Ln; t‘u} “ .4 s .. l ,u L1 Y-(‘K' .___ ‘1',- J .4; ' \' is v ' ’.' .. LL. «f r. ‘ ‘ 4 0 . - f I I Q a O ‘ \ q u .- 0 '; .‘ ’~. (x; c- .. I 3 -. :1 "u I . , ‘I,'.~ . .3: ‘g o "l ‘ a ’ ,“,’ ..‘ ,3, \' 'Q a ...-7 I \\ at n ’ ~. 3 "Q Jul - 1. .0... U') (2| L4,. .5 II IL‘ ‘1 {t"_ }:L.- 1‘: i ‘1‘, .i l tJALI-IJ 11“,...5‘3‘,“ .= t ...tI—iL-l it .1 5) ...:J‘, '. s ~ v ,. I. \ I L. g x s . o . _ "o~r ‘w" m4 3 ”-5. s, I! ' -J 1 “3‘ .. -) '3 .. \I.‘ 1 n- ‘1’“; Y03 If a : ift‘e were to be formed.to make Lb e of the fa ts brought ca: 3;? health surwu3, Eculd vou be willing to tUie 8n ggfigvg part in Eh, H3rk oi such a caumiteae? (890“ call on people to Jivé"then the fag-Tn Of the “T"VG“3 ”Ecziza 1339 J me-tings write up reports, etcg In genfiflli, 1‘" 1r 3'3 an} ) .. Kilns-n val-I'M {Wazch for: "on zc“iaam f3“ healtfi act’on" to “ftd up" kinds oz react;: . ‘3 73‘ ~ ) xv l 0 ~ - - Q i ' 7 ’ 3 ' 7 '3 “-fll' ,*' "A!“ «#1 (3% I v~,,,~ -~-- '1: q Jq-P. .. ,..a 3 3 .~ _ ., ..,.. ‘ 1 . ..1 . .. R l‘ .. "‘ 3 ‘n(.-"!.- \1‘3_,. -};. -_ V sir r\ . ‘ ’4’” “ ""' “Q 3 ""3“ {- 43m. Cu.- ,‘JU 'ir.V Ulwlx; 5321.2”.- w-'.'.' 1.- 1L". 3;»..‘v. w 3,.-. I. “I'll. 53‘2"“: x; K“ G1 1.1!.73-1‘. u‘fi r' "“~‘" 11' r IJQ .1. ( II I up. 3 f.- n ’ an} _. _,‘ i -- L fu'VA‘ v. ' ‘ ‘ I “3" ,4- V ‘r 1'— .9-~.q\-; ’/ vi.» - 3..“ av~ \J. ' ‘33") at». - - . - 3. '. L(-1: u ,-,.L.-.' .- ya)” a g. 32‘. ..z ..‘--UJC' ~. 3, " \ .1 'H _- I . 1 -“I- Htrmwru-r-p -.~r.,'-'-—u.p~:-‘r :3 / ' u . -‘ 1"" '7' '1': k .i i 5.4. t. ... 1.. 3‘ , ..’ .. , . IF...10"M*“¢"u.h“-‘I~'~I -.cofi"-¢u M"--- ~.-¢ ' A ' )- ~ 1 :~ . - - 1‘ i ' n’.". “ "’1 r“: r} ‘7 ,n. 3,": '- a ..; r’ j 'l Eut) A. .‘o ... M». Styx-”n1 ”'1... x." '4. .',€_.._‘ ‘ I“vwfl . -‘~ -91“ ’- ’ M‘-‘.w~ I‘m.“ n .. 3 . . .. ‘0 _ 4 n _- v - 9 ' 1- - ‘ . - . s , ‘ -\ v.' 1 -1 lg; ‘n‘hc, ‘ .'.fl .3 .‘3') . .... 1‘" 1‘ .. , ,/(‘ g - --‘ - ,— ., . ~. v... r, 3 . 3' :43. .3 . ‘f‘ | J - ‘1 J 3.14.3 If". H .5. . .,. LIQ' “LU J...) "3 ..'1.: L'n‘. LAIU “I ’.“ K ‘. 1- '. D ;' 'A .J‘~ “‘2‘ '“I 4 a. r. '1 ~- ‘ i‘ .5- I]. F‘ 1‘... i ' .H‘ T ‘ _ -" u i. ‘ A .. .f _ -_ L n- 3|". L- I .'3 - ‘ . .. 3. ‘1 \v . a -_\1 J I \ a l , \ HICHIGQN STQTE UNIV. LIBRRRIES 31293102733551