The Stage of Developing the Question - Essay Example

Developing the question The stage of developing the question is equivalent to the staging or defining directions in practice. It means clarifying the question after it s formulation, exposing some initial assumptions just like in practice. The literature review To carry out applied research, it is necessary to consult other sources of information. This consultation of other sources (often written) is called the literature review. You can find these sources in libraries and other social organizations and even private collections. Uses of the Literature review It is useful in developing the question as 1) connects the research question to the theory 2) Identifies previous research 3) gives direction to the research project But the literature review is also useful for guiding practice. 1. Connecting the research question to theory In any research a connection to theory has to be made. It is a straight forward exercise in pure research but not in applied research. For example to investigate causes of spousal abuse it is important to learn the theories of human behavior which are found in existing literature. Here are examples of the use of literature reviews to link different types of social work to theory. a) Practice evaluations You will need to know the theoretical base that guides the use of a particular intervention when evaluating your practice. E.g. if using positive reinforcement to teach parents disciplining skills you need to be aware of the theory behind positive reinforcement which is behavior theory and so consult literature on behavior theory. b) Program evaluations For each form of program evaluation i.e. summative, formative or cost-benefit you will need to consider how the research question links to theory. E.g. to verify whether the agency meets its goals in providing services to homebound elderly you will consult literature to find the theoretical basis for giving this type of care and also check other studies that have been done in this area. Example: Schaffer and Lia-Hoagberg (1994) assessed the personal, family, and provider rewards and costs low-income pregnant women experienced in obtaining prenatal care. Exchange theory provided the theoretical framework for this evaluation of prenatal services. The theory suggests that humans avoid relationships, interactions, and feeling states that are dissatisfying or costly and seek out experience that are gratifying, pleasurable, or rewarding. b) Needs assessment To determine the needs for a program a literature review can provide a theoretical; base e.g. to make a needs assessment for the number of homeless women and children in the community, the literature review can give a theoretical perspective and context. 2. Identifying previous research It is useful to know if the question you are investigating has been answered before. If it is recent then its results will still be relevant so you may have to reformulate your question and if it is not recent, you can repeat the study. A program evaluation may have been done in a program similar to yours and your evaluation would not give any new information. 3. Giving direction to the research project A literature review does not only help develop the question but also give direction and guidance to the research project. E.g. Literature review can show researchers how set up comparison groups in similar projects or give ideas how to select samples. This way the researcher can save a lot of time. Accessing information There are different sources of literature- libraries, government agencies, public and private organizations all have specialized collections. There is also a lot of material on the internet. Libraries Even the growth of the internet has not replaces the role of libraries. Libraries are still popular because not all information on the internet is free. Libraries offer tools for research like indexes, abstracts and databases. Library catalogs These are computerized catalog database that show what is available in a library. You can search online by authors name, the title or subject keywords or a combination of these. The database will give a list of close matches and you can sift through them. An easy way to use the library catalog is to use a “subject heading”. This can turn up books, dictionaries, bibliographies and journals which you can then narrow down. Government documents: They are useful to social research as they contain census data, congressional proceedings, agency information and statistics. Some libraries have a separate online system to access government documents. A lot of government information is increasingly available online. Indexes and databases Can either be on the web or in print or CD-ROM in the library. Helps to identify specific articles, dissertations, chapters in books, books and documents not limited to just you library. Such indexes can offer a literature review in your area of research which can lead to further relevant resources. Some indexes/ databases are specific in subject area or type of in the library o another library. Some databases can give full access to full text of articles citations, or links to full text. To search an electronic database is the same as searching a library catalog- you will use keywords and can also limit the search with some parameters. Access to web-based library resources Resources that libraries subscribe to electronically are usually covered by licence agreements that limit access to valid members of the library e.g. university faculty, staff and students. You will need a library card which will allow you access from any location to the library’s web resources. Without a library card however you will to visit t the library and search using their computers to search web-based subscriptions. Interlibrary loan Library indexes and databases give information on resources not available in your library or on the web. The interlibrary service will enable you to get these resources. Libraries may charge the full cost to the client or subsidize it. It is a process that takes time so you will have to start early. Library staff Professionally trines staff can save you time in your search. Make a habit of asking for their help. 2. The World Wide Web The World Wide Web contains lots of information that is useful to social work researchers and increasing access to international information. The disadvantage is that it is time consuming sifting through all of it much of which will not be useful. It is made worse by search engines that lack precision or engines that put commercial interests first so these sites appear at the top. It is wise to bookmark any useful site you come across as it is hard to remember how you made the search in the first place. You can also use gateway sites that organize well –maintained access to web sites of interest to a specialized audience e.g. social work gateways by schools and departments of social work, libraries and organizations of social work. Example Vernon and lynch (2000) provide guidance on the use of the web in research. A summary 1. State the topic, limit the range, and list all the relevant synonyms and keywords. Use background sources in appendix A for help with terminology and scope. 2. using the computer catalog in you library, use the keywords to access potential sources of information. 3. When you access a relevant item, look at its subject headings and use these words in the subject headings category to find similar items 4. If relevant, locate other materials in the government documents 5. Use your keywords to search the online databases available through your library 6. Use a print index if you have not found enough results in databases 7. From your results, try to develop a literature review on your research topic. Consult the list of references for further resources. 8. Use gateway sites on the web to track down other material. Verify the source of web-based information and evaluate it for accuracy, currency, and integrity 9. If material that you have identified is not available locally or no the web, use interlibrary loan through your local library. 10. Ask reference librarian for other ideas and help, if your need it, for each of these steps. WRITING THE LITERATURE REVIEW It should be completer before other stages of research are undertaken. It places the research in its historical and theoretical context. It gives the research background and the relationship between the study and past studies in the same area. The literature review will also identify trends and debates in the existing literature. (Van Rooyen, 1996) identifies issues to consider in the constructing the literature review. Issues to consider when constructing a literature review (van Rooyen 1996) (56) Cite only research that you find specifically pertinent to the current study; be selective. Avoid reviewing or referring to sections of articles or texts that are not related to your study Discuss and evaluate the literature you have selected Show the logical continuity between existing literature and your study Identify controversial issues or differences in the literature and your study If there is a choice, cite the more recent literature, unless the older citations are needed for additional perspective Write the literature review in the past tense Refer to published studies for examples of literature reviews UNITS OF ANALYSIS After the literature review the question is developed further. The first step is to identify the unit of analysis i.e. what or whom is being studied. There are 3 types: 1. individuals 2. groups 3. social artefacts Individuals: These are the most common units of analysis often analysed to explain how the social group functions e.g. for a needs assessment, for a community youth center, would interview individual youths to assess their needs. Groups: These could be families, organizations and communities e.g. a program impact assessment on family cohesion, he family will be the unit of analysis though it is the individuals who will be studied. Social artefacts: Are behaviors or products resulting from human activity; e.g. books, divorces, birth practices, or ethical violations E.g. If asked by your state NASW chapter to investigate unethical social work practice. In your study you would look at those charged with such behavior whether they are B.S.W.s or M.S.W.s, the field they practice in and so on. The unit of analysis is unethical social work practice (57). LEVELS OF MEASUREMENT The next step in developing the question is to outline the level of measurement. It is the extent to which a variable can be quantified to be eventually processed statistically. Quantification means giving a number to a variable and it depends on how the variable will be operationalized. E.g. to measure depression you can use the Generalized Contentment Scale or count how many hours the client sleeps. There are four levels of measurement 1. Nominal 2. Ordinal 3. Interval 4. Ratio Nominal: Phenomena are grouped based on similarity or difference e.g. ethnicity, gender, marital status; numbers that are assigned to nominal groups but have no inherent meaning themselves. E.g. it could be 1 for Hispanic And 2 for African-American The Numbers can be reversed without lose of meaning. Ordinal: Ordinal measures classify observations into mutually exclusive categories that an inherent order to them. Often used to measure attitude. A few examples could be Strongly Agree Disagree Strongly disagree They are in sequence and numbered 1-4 but the distance between them is no equal. Levels of Measurement table 4.1 Level of measurement Definition Example Nominal Data are assigned to categories based on similarity or difference Ethnicity, marital status Yes/no response Ordinal Data are sequenced in some order Many attitude and opinion questions Interval Data are sequenced in some order, and the distances between the different points are equal IQ, GRE scores Ratio Data are sequenced in some order, the distances between the different points are equal, and each value reflects an absolute magnitude. The zero point reflects an absence of the value. Years of age, number of children, miles to place of employment Interval: This measure groups observation into mutually exclusive categories with order and equal spacing between the categories. The equal distance between the categories makes its different from ordinal measures e.g. the IQ test; the difference between an IQ of 120 and 130 is the same as the difference between110 and 120.However, this measure does not mean that one value is greater than another e.g. we cannot say that an IQ of 160 is twice that of IQ 80. Ratio: Are similar to interval measures but show the absolute level of the value i.e. to say at zero it means the value is absent or did not occur e.g. income, years of education, number of times behavior occurs. Most variables can be measured at different levels e.g. anger can be defines to measure at different level 1. Nominal: Do you think your child is angry? = yes /No 2. Ordinal: To what extent do you think your child is angry? = very aggressive =aggressive =not aggressive 3. Interval: if anger is measured as one factor in a personality test 4. Ratio: if anger is defined in behavioral terms e.g. how many times the child hit another in an hour. NAMING THE VARIABLES AND VALUES A variable is the quality of a phenomenon and it varies e.g. in social work research –income, ethnicity and stress level. The qualities vary in quantity and the different quantities are the values. Values are the potential answers to questions e.g. values for income could be Under $ 15,000/year $ 15000 - 19999/year $20000 - 24999/year $25000 - 29999/year $30000 and over a year Values for ethnicity: White (non-Hispanic) Hispanic African American Native –American Other Values for stress: High, medium, low Variables and values differ from study to study. E.g. A survey to assess the need for a day care centre for developmentally delayed preschoolers, one variable may be income to see what extent parents could pay for the service. In rural Kentucky you would anticipate that incomes will be low so the values in the survey will be low but the same study in Santa Barbara, California the same categorization might be too low and would need to add higher income levels. For the same survey ethnicity may be considered a factor that influence service and can be included in the study. Same study will be different if done in New Mexico and South Africa; completely different values will be used for ethnicity. One problem with naming values in this way is that information is lost. E.g. clustering individuals in a group like “native American” leads to the loss of potentially critical information: like the difference between tribes, places of residence (on reserve, rural or urban0 etc. This is where the interpretist approach is useful when unsure of the nature of the values to be included. In interpretist studies the values and variables are not named before the study but emerge from the study. NOTE: Be ware of reductionism when deciding variables to include in a study. This is the extreme limitation of the kinds and numbers of variables that might explain or account for broad types of behavior. Reductionism is common with positivist approach where all the variables are named before the study. Time and money constraints often force researchers to consider a limited set of factors in a problem. It is not a problem until they claim that the factors studied are the only ones that influence the problem, this would amount to reductionism. It is important to acknowledge the possibility of other explanations or factors in discussing your results. Example. A study on spousal abuse, many factors can be considered like economic factors, biological family dynamics, psychological and so on. If you focus only the economic and claim that to be the only cause then that is reductionism (61). THE RELATIONSHIP OF VARIABLES Variables have different functions in different research studies. Dependent and independent variables roles are the major distinction. In explanatory study the specific variables are identified as contributing to specific outcomes i.e. to establish causality. In descriptive studies like needs assessment independent and dependent variables are not identified. Interpretist studies do not identify independent and dependent variables as they are not caused directly with causality. Independent variable: Is the variable that can affect other factors in the research e.g. studying the impact of social isolation on child sexual abuse – so social isolation is the independent variable. In a program evaluation it is the program Dependent variable: It is the resulting variable that has been affected by the independent variable e.g. summative program evaluation to see if program goals are being met – it would be the goals that would be the dependent variables. In a study of factors leading to child sexual abuse in a community, the child sexual abuse is the dependent variable. Itzhaky and York (1995) studied the impact of apartment building committees in Israel. The house committees employed tenants as maintenance workers and administrators and also introduces professional workers. They chose dependent variable as whether social activity of any kind for the residents took place. They identified 3 sets of variables: 1. the background of the families living in the building and the community typology2. The functioning of the committee, how it was set up, and how many meetings took place and how dues were collected and 3. Whether a community social worker was in contact with the committee (62). A study of factors leading can have more than one independent and dependent variables. E.g. In the child sexual abuse study, income level would be an independent variable and different kinds of abuse will be different dependent variables. The literature review is important in identifying the dependent and independent variables. Variables are not fixed as dependent or independent but depend on the study’s purpose and context. DEFINING AND OPERATIONALIZING THE VARIABLES Clear definition of variables is important in the positivist approach so they can be measured but not so much in the interpretist approach where definitions of concepts/ variables appear as the topic is explored. However the focus of an interpretist study still needs to be clearly defined e.g. a study of beliefs about mental health illness; the researcher has to clearly define “mental illness” (62). Example Margareta Hyden (1994), in her interpretive study of woman battering in Sweden, clearly defined at the outset of her research the definition of assault or aggravated assault where the woman was the victim and the husband the perpetrator. Her study involved interviewing women about their experiences and how they made sense of violent action within their marriages. What emerged were different definitions and interpretations of the concept of marriage (63). Lots off variables used in social work can be vague and interpretation may depend on who is using them. It is important to define them to avoid confusion e.g. “independent living” may mean quite different things to a client and social worker. The client may think it means “living in his apartment with no supervision” while the social worker, might think “living in his apartment with close supervision” in which case we have to define “supervision”. Another issue to consider in defining variables is the differences from culture to culture. This is more so when studying unfamiliar cultures. The interpretive approach is advised as definitions will emerge from it. The literature review will help define the variables. OPERATIONALIZATIONS In the positivist approach after defining the variables is to operationalize them i.e. state how the variables are to be measured. This step is not necessary with this step, since the purpose of the study is to understand different dimensions of the variable. Operationalizing is easy when variables have been formally defined. It can be challenging though because concepts like depression can be hard to define. It is useful to look at the literature and see how it has been done by others. A variable in social work may be measured in many different ways e.g. depression can be measured by the Generalized Contentment Scale (Hudson 1990) or the Depression Self-Rating Scale for children (Birleson, 1981).These two measures can be useful or not depending on what aspect of depression you want to measure. Example Cox (1995) compared the experiences of African American and white caregivers of dementia patients. Among the variables operationalized for the study were 1) attitudes to ward giving care giving, assessed by the extent to which elderly people should expect to receive care from relatives and the extent to which relatives should be expected to provide such care; 2) patient status, assessed according to the level of cognitive functioning, disruptive behavior, social functioning, and physical functioning; and 3) caregiver stress, measured according to the caregiver’s sense that his or her activities were restricted and relationships strained.(64.) Defining and operationalizing goals and activities Special attention to summative program evaluations: It is the evaluation social workers make to see if a program has met its goal. The positivist approach is recommended and the program goals and activities have to be defined and operationalized. What is meant by “goal” and “activity” because these terms are used differently by different people. Example: The goals of Adolescent Family Life program may be to reduce rate of high-risk babies born to adolescents and rate of child abuse and neglect among teenage parents. Activities can be providing prenatal care and parenting classes to teenage parents. This is how we operationalize these goals and activities: Goal 1 = reduce rate of high- risk babies born to adolescents= low birth Defined as low birth weight or premature infants Defined as those 18 years and under After we have to operationalize low birth as under 5 ½ pounds and premature as born after 32 weeks or less of pregnancy. ETHICAL ISSUES IN DEVELOPING THE QUESTION 1. Giving credit to contributors To authors used in the literature review in the final research report 2. including relevant variables It is important to include all the important variables it is especially important in positivist research where variables are clearly defined before the research is started. 3. Avoiding reductionism 4. Reductionism occurs when the researcher looks at one variable and claims it as the only cause responsible for an outcome. E.g. in a program evaluation, reductionism would lead to consideration of only those variables associated with the program and no others. The other factors may have even more impact than the program in question itself. Reductionism is more common with positivist approach. E.g. in evaluating a program to enhance self – esteem among high school drop-outs, you include variables like length of time in the program but not other outside factors that may influence self-esteem like participating in a local sports activity. The other activity may have more impact but including them may jeopardize the efficiency of the program (66). HUMAN DIVERSITY ISSUES IN DEVELOPING THE QUESTION Human diversity issues should be considered to avoid biases against some groups. Example: McLaughlin and Braun (1998) discussed the differences between individualist and collectivist orientations in health care decisions. Americans tend to have individualist orientations, whereas Asians and Pacific Islanders tend to have collectivist orientations, meaning that many decisions are made by families and groups rather than by the individual. This is especially true concerning health care decisions. The authors note that the collectivist orientation is not well represented in the research and literature. They recommend that universities and health care settings provide education and training to promote cross—cultural practice and sensitivity (67). There can be bias in literature because most research has been undertaken by white middle class men and women. This overrepresentation of these groups is a bias. If research questions were developed by other groups they would be different. Cultural factors influence each process in developing the question e.g. how a variable is defined in influence by the culture in which it exists. Example: cultural definitions Lowery (1998) described American Indian perspectives on addiction and recovery reaching beyond simply an intellectual understanding of “healing the spirit”. The article was organized to reflect the movement between “spirit and science” and included four concepts 1. balance and wellness 2. the colonization experience and addiction as a crisis of the spirit 3. issues of abuse, including sexual abuse, and 4. a time of healing illustrated by a Lakota commemorative event. The author suggests that social workers must explore their definitions and perspectives on healing to enhance both their practice and research (67). E.g. the example on independent living may mean 1. culture 1 = living the family and employed outside 2. culture 2 = living with the family and being married Read More