Needs Assessments
Leslie M. Tutty and Michael A. Rothery
N eeds assessmen ts a re a for m of resea rch conduclcd to gather information about the needs of a population o r group in a co mmunity. One of the more practical types of research, needs assessments are used to develop new services or to evaluate the relevance of exist ing programs. They may also be used to establish a need to rev ise or create policy.
Th is chapter begins with a definition o[ needs assessment, how we define " needs:' and how we determine who to ask abou t needs. Common methodological approaches to n eeds assessments are described and evaluated using examples p rim arily from the social work literature. The benefits of triangulation, or using more than one source or method of gathering information, are presented, followed by a discussion of who shou ld digest and weigh info rm at i.on about ne eds once the information is ga thered. Finally, we conside r th e importance of developing a plan to implemen t recommendat ions so that th e work of assessing needs is used to clients' benefits, not relegated to the shelves occupied by other dusty and neglected reports.
What Is a Needs Assessment?
Needs assessments have not changed much over the years. In 1982, Kuh (cited in Stabb, 1995) listed five general purposes commo nly served by needs assessment research that remain relevant today:
l. Monitoring stakeholders' perceptions of various issues, which can guide the devel-opment of new programs or policies
2. Justifying existin g policies o r programs
3. Assessing client sa tisfacti on with services
4. Selecting the most desirable program or policy from several alternatives
5. Determining if needs have been met, a purpose closely akin to program eva luation
Two key q uestions are addressed when needs <1ssessments are undertaken: "Wh o?" and " IIow?" The "who" questio n requires the researcher to be dear abo ut the membersh ip of the group whose needs are to be assessed. Often, a study entails gathering information from a variety of respondents, from individuals who may never have been clients to those
149
150 PART II • QUAII.TITATIVE APPROACHES: TYPES OF STUDIES
receiving multiple services. In almost every case, however, at least one set of respondents will be the individuals who are most immediately affected by gaps in services or supports, rather than relying so lely on the opinions of service providers, academics, or funde rs .
The "how" question addresses th e methods used to gathe r informa tion from the group whose needs are of interest. These are not unique; rather, needs assessments borrow familiar techniques such as surveys, interviews, and focus groups, all of which are high-lighted in other chapters in this book. Quant itative methods such as surveys or standard-ized m easures may be used, as may q ualitative m ethods such as in-depth individ ual interviews or focus groups. Combinations of both arc increasingly popular since each method has its advantages and limitations.
Defining Need When we invoke the concept of needs, we may easily assume that we share with others a common understa nding of what it is we are talking about. However, it is worthwhile look-ing more closely at the defini tion of the te rm since usefu l characteristics and di stinctions are highlighted when we do so.
The concept of need is not new: Researchers have been defining and red efining the term for decades. Stabb (1995) distinguishes between met and zmmet needs. "Met needs are necessary or desirable co nditions that alrea dy exist in act ua li ty. Unmet needs arise when there is a discrepancy between desirable conditions and current actuality" (p. 52). Both met and unmet needs could conceivably become the focus of needs assess-m ent research, although unm et needs will be the main concern in the vast majority of cases.
A different distinction (perhaps m ore usefu l for our purposes) is provided by Witkin and Altschuld (1995), who define a need as "a discrepancy or gap between 'what is,' or the present state of affairs and vhat should be; o r a desi red state of affairs" (p. 4). In this an alysis, needs equate with unmet needs, the most common fo cus fo r needs assessment research. Revere, Berkow itz, Carter, an d Ferguson (1996) add the sugges tion that need is defined by "community values, [a nd is) amenable to change" (p. 5).
From these perspectives (and wi th reference to considerations introd uced earlier), a needs assessment gathers informatio n about gaps between real and ideal conditions, the reasons that these gaps ex ist, and what can be done about them , all wi thin the con text of t.he beliefs of the community and available resources for change.
Another distinction introduces the question of degree. Some needs are stronger or more important than others. f undamental needs with relevance to people's sur vival, safety, or basic comforts are not the same as "wants" or less compell ing needs. A social work professor's desire for a week in Mexico as a break from winter is qualitatively very different from a homeless person's need for food and shelter in the face of the same cold conditions. While it is often d ifficult to draw the line between rela tively impor ta nt needs and less important wants, it is still important to do so. Needs assessments are focused on needs that affect ind ividuals' abilities to function well in important areas of their lives. Wants associated with perceived quality of life (b ut not to the same extent with life's real essentials) are more lhe purview of market research.
Social workers generally find Maslow's (1970) hierarchy of needs useful when consid-ering the needs and priorities of their clients. It is also a framework that can inform needs assessments. Maslow's five levels of need are physical and life-sustaining needs (such as air, water, food, warmth, and elimination of bodily wastes), physical safety (e.g., protec-tion from physical attack and disease), love and support, self- esteem an d self-wo rth, and self- realizat ion (e.g., needs to be productive and creative) . Maslow con tended that these
CH APTER 9 • N EEOS A SSESSMENTS 151
basic needs must be attended to before attempting to address higher level needs (or "wants"). Needs assessments can gather information relevant to any one or more of these five levels, b u t the hierarchy of priorities prov ides useful cr iteria fo r deci d ing on what to focus first in data collection and recorn mending changes.
Finally, some authors argue that once an "expressed need" is verbalized, it becomes a want or a demand (Stabb, 1995) . This is not the same as differentia ting needs from wants o n the basis of the strength of the potential impact on someone's well -being and is prob-ably less useful for our purposes. However, a related point is noteworthy: Verbal demands are not always the d irec t expression of need. Just beca use someone exp resses a want d oes not mean that it represen ts a need. Thus, in needs assessments, it is important to gather information from members of a population beyond those publicly advocating for specific d emands.
Who Do We Ask About Needs? The term stakeholders is often used to refer to clients or potential clients or the people who actually n-perience the need thal is being studied. However, Revere and colleagues (1996) suggest broaden ing th e d efinition to refer to "service providers and m a nagem ent, com-munity members, certain politicians, the funding source, business/trade associations and the actual research •..vorkers" (p. 7) since each of these has a vested inte rest in the study and its o utcomes. Th is flex [ble use of the term is helpful, suggestin g a range of potential sources of data and recognizing that needs assessments have ramifications for people beyond those normally surveyed .
Needs assessmen ts tra ditiona lly look to three groups as sources of data: the target group (i.e., clients or potential clients), key informants such as community leaders or service providers, or a sample of aJI members of the relevan t comm unity. Each is described in more detail below.
The target group or populatio11 comprises the very individuals about whom we arc con-cerned and whose needs we w ish to assess. Common sense suggests that these are the voices we most wish to listen to in o ur q uest t o gather the best an d most current infor-mation. However, engaging with individuals to encourage them to share their needs and op inions is not always easy. Highly d isadvantaged, socially m arginalized individuals and grou ps, the typ ical focus of social workers' interventions, are nol always accustomed to being asked their opinions and may not easily articulate their needs to a researcher when inv ited to d o so. Furthermore, they may have understandable reasons fo r n ot trusting members of those who have mo re powe r in soc iety, a group to which researchers belong. Consider the homeless as an example, especially the subpopulation that has been diag-nosed with psychiatric disorders. With any such group, the researcher ca n not simply approach and invite them to en umerate Lheir needs. Strategies (and time) for building trust, rapport, and for encouraging engagement in the research process are prerequisites for successful data gather in g.
McKi ll ip (1998) defines another gro up serving as a common source of data, key infor-mants, as "opportunistically connected individuals with the knowledge and ability to repor t on communi ty needs. Key in fo rmants arc lawyers, judges, physicians, min isters, minority group leaders, and service providers who are aware of the need and services per-ceived as important by a community" (pp. 272- 273). An advantage of gathering data from key informa nts is that they may have a broad er knowledge of serv ices available in the community than the target population, and they may be better at articulating what needs must be effectively addressed. One disadvantage is that key informants sometimes have a vested interes t in developin g new services or preserving esta blished resou rces even
152 PART II • Q UANTITA IIVE A PPROACHES: T YPES OF STUOIES
though they arc less than adequate (we all develop loyalties, and these can affect our judg-ment). McKfllip (1998) also notes that key informants may underesti1nate the willingness of m embers of the target population to participate in programs while overestimating the extent of the problems.
The third group, community members, comprises th e entire citizen ry of a comm unity. This group encompasses members of the target population but also includes those not directly affected by these needs. App roaching community members for information has the advantage of identifying how broadly based the needs are, rather than assuming that they are restricted to the ta rget population. lt also offers the opportunities to lea rn about how needs (and the strategies Lo ameliorate them) are perceived in the commu ni ty at large and to think about how that v.rill affect efforts to implement changes. A disadvan-tage, though, is that community members m ay be relatively t111awarc of the needs of its more marginalized citizens.
In summary, each of th ese groups may b e the focus of the needs assessmen t methods documented in the next several sections. The choice of whom to engage may be based on access to the group or limitatio ns of tin1e and resources. If possible, representation from each of the target population, key commu ni ty stakeh olders, and members of the general public is worth considering as each provides valuable but somewhat different information.
Methods of Needs Assessment
As mentioned prev iously, one ca n co nduct needs ass essm ents u sing a variety of strategies. We will d£scuss methods in lwo broad catego ries, quantitative and qualitative. Quantitative m ethods gather data th at are tran slated in to numerical form and described using statistics. Using such methods, it is possible, for example, to conclude that in a sam-ple of 102 shelter residents, 70.5% of these women abused by intimate partners were abused themselves as childre n and described 73.7% of their pa r tners as also having been abused (Tutty & Rothery, 2002). Such high proportions may be interpreted as suggesting the need for early in tervention with children in sh elters in the hope of preventing the cycle of violence from affecting a new generation.
Providing statistics about the extent of a need can be a powerful method of raising awareness of the severity of gaps in services. The section on quantilative needs assessment will describe three such methods: surveys, s tandardized needs assessment measures, and using existing statistica l databases.
In cont rast, qualitative needs assessments ask questions that are more op en-ended and allow the research in fo rma nt to describe in detail the complexities o f the issues at hand. for example, a qualital ive needs assessment conducting in terviews with another group of ,63 abused women residing in a shelter noted that providing for their basic needs such as safety and food was of great impo rtance (Tutty, Weaver, & Ro thery, 1999) . However, some women expressed concerns about the fact that a few residents were difficult to live with, and some mothers did not m anage lhcir children's aggressive behavior or ignored them. These results suggest a somewhat different focus for intervention by crisis counselors and th e need to p rovide parenting progra ms for some residents.
Results from qualitative needs assessments often lack s ta tistical dat a that could convey the extent of the problem, but they tend to be rich in detail that conveys the complexities and uniqueness of the experiences of different individuals. T he quali ta tive needs assess-ments methods described in the chapter include interviews (either face-to-face or by tele-
~!1~ll~h f8EU~ KrGUp~. nomtm1 groupg, ~nd t CMh halt meetings.
CHAPTER 9 • NEEDS ASSESSMENTS 153
Quantitative Methods of Needs Assessment Surveys Allhough surveys may ask open-ended qualitative questions, the great majority are devel-oped for quantitative data analysis. Quantitatively oriented surveys, particularly those employing questionnaires, are the most freq uent method of assessing needs. The tasks involved in developing a survey to assess needs are ident ical to those undertaken when surveys arc developed for other purposes, so they will not be detailed here. The major steps involve
1. D eciding who Lo survey (e.g., target groups, key info rmants)
2. Selecting a method of sampling (e.g., random or systematic sample)
3. Determining the content of items (through reviewing the literature or holding focus groups with key informants, as only two examples)
4. Ch oosing what type of question to use (e.g., open-ended, multiple choice, or scaled with respect to the extent of agreement)
5. Selecting a method of distribution (e.g., the Internet, mail, or telephone)
The advantages of surveys include the ease and flexibility with which they can be administered compared to other methods and th e relative lack of expense to collect a con-siderable amount of data. Disadvantages include the extent to which a set questionnaire can predetermine the issues that respondents address and the consequent danger of not hearing about needs that would emerge in a more open-ended process.
With such risks in mind, Witkin and Altschuld (1995) recommend being cautious about assuming that a written questionnaire is the most app ropriate tool wh en considering con-ducting a needs assessmen t. While a questionnaire can be an important tool, they suggest that it should not be used until after more exploratory methods have been employed to ensure that the factors measured by questionnaire items are as well chosen as possible.
Furthermore, some cultural groups find surveys strange or difficult (especially if English is not one's first language) and respond negatively to them. Weaver (1997), for example, described a questionnaire developed to assess the needs of an off-reservation Native American community in an urban area. A large numb er of qu estionnaires were mailed out, with virtually no returns. The alternative of a qualitative approach including focus groups and individual interviews was adop ted with considerably greater success.
An example of a needs assessment that employed survey methods more appropriately is Brennan Homiak and Singletary's (2007) study that surveyed Christian clergy members from 15 denomina tions in central Texas with respect Lo their perceptions of the number in their congregation experiencing intimate partner violence and what clergy needed to better address this serious concern. Of the 100 surveys mailed, 44 were returned, a some-what low but not unusual return rate for mailed surveys.
The clergy members estimated that less than IOo/o of thei r congregation members experienced partner violence–low when compared to incidence studies in Texas that cited lifetime rates of 47%. Only about one third of the clergy had received domestic violence-specific training; they were more likely to have resource materials in their churches and were familiar with local agencies and shelters for abused women. While a small proportion of the clergy considered themselves very equipped to counsel victims of domestic violence or make referrals, the majority did not. The authors recommend th at social workers take the lead in offering trai ning to assist the cl ergy in promoting violence-free congregations.
154 PART II • Q uANTilATIVE A PPROACHES: TYPES OF STUDIE~
As mentioned prev iously, surveys may usc both ch eck-lis t type, predetermined resp onses and open -ended questions that allow for m ore context ual detai led responses and arc analyzed using qualitat ivc m ethods. A recent exa mple of using open-ended ques-tions is a survey with 206 agency-based social work field instructors, querying their initial awareness, personal and professional needs, and field issues that arose in response to th e World Trade Ce nter disaster of Septem ber lith, 2001. The field instructors had clearly been weary but retained sensitivity to studen t and client n eeds. The res ults suggest th e imp orta nce of develop ing an integrated crisis p lan to better lin k the school, students, and field instructors in the event of future disasters.
Standardized Needs Assessment Measures A relatively n ew needs assessme11t methodology entails developin g stand ardized measu res to assess th e needs of a specific popu lat ion group. For exampl e, Wancata and colleagues (2006) initially used focus groups and in depth individ ual interviews to develop a mea-sure comprising 18 common p roblems exp erienced by caregivers o f adults diagnosed with schizophrenia. The difficulties were translated into items such as "not enough infor-mation on the illness, its symptom s and course," " fear of stigmatization and discrimina -t io n ," and " burnout or illness of the carer."
Using such a measure in other needs assessment research has the advantage of building on the work that has gone into identifying and conceptualizing potentially impo rtan t n eeds and of using a meas ure for whi ch reliabil ity and validi ty will often have been estab-lished. A poss ible disadvan tage is that needs proven relevan t to caregivers of adults diag-nosed 'l'ith schizophrenia in one location may not ha ve the same importance in others. Conversely, items about other needs tha t are important in a new loca le may be m issing from the standardized measure.
Using Existing Statistical Information Another quantitative method of con du cting needs assessments is using da ta that h ave been previously collected. Existing data may be available in agency fi les or governmen t data banks, for example. Such secondary analyses have the advantage of sparing researchers the tim e and expense of gathering new data. A d isadvantage is that one is lim ited to data th at som eone else considered worth gath ering, and potentially important variables may be absent or may need to be inferred indirectly from the data that were recorded.
Review ing case fil es can be challenging. As a follow-up to a previousl y completed study on the o utcomes of a specialized domestic violence court (Tutty, McNichol, & Christe nsen, 2008), we are using district attorney files to collect a number of variables, including the demographic cha racteristics of tbe accused and the victim, whe ther the victims testifi ed or provided a victims impact statement to the court, and hov,, the trial was resolved.
In contested cases, the files can b e very large, literally inches thick! The fi les are created for the crimina l justice system, not researche rs, so there is no consistent organi zation. As such, collecting data from one file can take several hours. Considerable information may not be recorded. Lawyers are not necessari ly as interested in demographic cha racteristics such as ethni city or age as most researchers are, and so lit tle of th is can be found in the files. Desp ite these challenges, if we are to evaluate the specialized courts, paper flle reviews are our only option to assess whether the courts are m eeting Lhe needs of both the victim s and th e accused.
The followi ng needs assessment used case records to assess wh ether the needs of abused and neglected children were adequately addressed by the child welfare int ervention . Tracy,
CHAPTER 9 • N EFDS A SSLSSMENTS 155
Green, and Bremseth (1 993) revi ewed case records of supportive services for abused and neglected children in one U.S. state. Five hundred child welfare cases were sampled to explore facto rs associated wit h decisions to offer one of two services, family preservation if children at risk were still at horne, or reunification fo r families with children who had been placed. The authors collected information on demographic variables, presenting problems, service history, service needs, services planned and provided, service characteristics, and serv ice outcomes. This en terprise, the aLlthors noted, consumed thousand s of hours.
The analysis uncovered significant stresses affect ing the chi ldren sampled, parental substance abuse, economic difficulties, and poor liv ing conditions, which were infre-qu en tly addressed in case plans, whi ch emphasized indications of child abuse. The authors conclutled that "there was little one-to -one d irect corresponden ce between the service need and the service offered" (Tracy e L al., L993, p. 26), raising serious questions about the quality of service pla nning (and the training of child welfare workers).
Qualitative Methods of Needs Assessment Qualitative needs assessment research may be conducted via individual interviews, small group discuss ions, or even large town hall meetings, each of which allow for more open exploration of issues than the q uantitative m ethods previously outl ined. Such studies tend to involve a greater t im e commitment from respondents but offer much more opportunity to identify and discuss issues in depth.
Individual Interviews
Face- to -face a nd telephone interviews are one method of gathering in-depth information about the needs of particular groups. Preparation involves thi nk ing through the purpose of th e interview, constru cti ng an in terview schedule, and train ing in terviewers (Vitkin & Ahsch uld, J 995 ) . When a goo d rapport develops between interviewer and respondent, the result can be disclosure of informatio n and ideas about sensitive issues that would not emerge when more formal, structured approaches are used. Also, in a more open-ended process, respond ents may identify needs that no one had anticipated.
The disadvantages of this approach inclu de the fact that it is notoriously la bor intensive. Interviews arc tim e-consuming to conduct, often lasting one to two hours, especially if asking individuals to reveal their personal stories. As a result, often only a relatively small sample of individuals may o r can realistica lly be interviewed. Training the interviewers also takes time, and the job of transcribing and analyzing interviews is normally a lengthy, com-plex task. The following needs assessment is an example of using face-to -face intervie ws.
In the past 30 years, intimate partner violence has become an issue of significant soci-etal concern, resulting in specialized justice and physical and m ental health sh ifts to more ad equately safeguard the women and children wh o are the primary vict ims. Yet certain ethnocultural groups, including immigrants and refugees, are underrepresented among those seeking assistance from formal supports such as the police and emergency shelters.
With respect to the question of what would constitute culturally appr opriate responses t o domestic violence, Pan ct al. (2006) conducted 120 face- to-face inte rviews with members of three ethnic communities in San Diego: Somali, Vielnamese, and Latino. The interv iews were provided in the appropri ate language, and within each cu ltural group, 10 women, 10 men, 10 boys, and 10 girls participated.
Beca use of tl1 e sensitive na ture of the issue, the topic of dom estic v iolence was intro-duced usi ng vignettes, rather than asking interviewees whether they had personally been abused. This allowed the respondents to speak about abuse in their culture in more gen-eral terms and to sugges t possible resolutions to the problem.
156 PART II • QUANTITITIVE APPROACHES: TYPES OF STUDIES
The analysis of the interviews highlighted six core issues, including "varying defmi-tions of violence, specific definitions of family harmony, strict gender rol es, varying con-flict resolution strateg ies, cultural ide ntity a nd spirituality" (Pan et al., 2006, p. 42) . The differing perspectives from the three ethnic communities suggested the need to develop diverse culture-specific services.
Foc u s Groups
Focus groups are rel a ti vely unstructured small group experiences, typically with about 8 to 12 participants. The group composition is usually homogeneous in that members share a particular experience or interest, like the members of what we described earlier as the target population. Focus group interviews typically take from one and a half to two and a half hours, and a number may be conducted for a given study.
Witkin and Altschuld ( 1995) summarize the process of a typical focus group. Initially, members hear a general statement of the purpose of the session and are given a question related to this purpose designed to elicit perceptions about important needs. Often, par-ticipants are asked to write down the ideas that the question stimulates and then to share them wi th the group. The leader typically writes ideas as they are shared, summarizing them and making sure th at there is agreement among members wi th what is bein g recorded. This process is then repeated with other predetermined questions.
Leadership is important to a focu s group's success, espec iall y since the re is no highly stru ctured agenda (except for the pos ing and answering questions aspect ). According to Witkin and Altschuld (1995), "The leader must be nonjudgmental, create a supportive group atmosphere, be able to keep the interview process going, be a good listener, and be alert to sense when a group is deviati ng from the prescribed question route in meaningful and non-meaningful directions" (pp. 172-173). These are by no m eans easy demands.
One advantage of group approaches over individual interviews can also be a disadvan-tage. Whi le participants do not have the same opportunity to explo re their own perceptions or experiences in depth as in individual interviews, a group approach can elicit information that would not emerge without the stimulus of i nterC~cting with others and rea cting to their ideas. When group discussions detour in innovative ways, th is may lead to original and cre-ative ideas. Brainstorming, or encouraging members to present any solution to a problem without prejudging it, is one way to encourage such innovation. Alternatively, withou t effec-tive facilitation , the groups may pmsue unproductive tan gents, and there is a heightened risk of interpersonal conflict detracting from the effectiveness with which research goals are p ursued. The fo ll owing study used focus group methodology.
A relatively new role for sociaJ wo rk graduates is working with senio rs and their ram ilies to assist these clients in a number of ways. Yet, how readily are social worke rs per-ceived as resources to this population? Naito-Chan, Damron-Rodriguez, and Simmons (2004) used focus groups to explore what skills social work practitioners need to ade-q uately address the Heeds of older persons and their families. The four focus groups included older adults and caregivers of older adults (consumers ) as well as providers of care and recent social work grad uates, both wo rking in gerontology settings.
Notably, the analysis highlighted that a numb er of the consumers had little under-sta ndin g of how social workers could assist them. Key among the consumer needs was resource finding, which th e consumers did understt~nd as a social worker role. However, other social work competencies such as assessments and case management were not men-tioned by the consumers. The results suggested the need for public education abo ut the roles of social wo rkers in the field of aging.
CHAPTER 9 • NEEDS AsSESSMENTS 157
Nom i nal Groups
An al ternate group approach to needs assessment has been developed (McKill ip, 1998). Nominal groups are more structu red than focus groups: The agenda allows group discus -sion but with a more consistent attention to the goal of achieving consensus about needs. Fewer needs assessments that use a nominal group approach can be found in the litera-ture. Although more than a decade old, the following study provides a model of nominal groups with respect to issu es that remain current.
It is com m on ly acknowledged that inter perso n al and so cial pro blems, whether at home, in the neighborhood, o r on school grounds, can seriously affect students' abi lity to learn. Gerdes and Benson ( 1995) used a nominal group process Lo assess problems expe-rienced by inner-city African American schoolchildren. The goal was to identify the most serious problems faced by students from their own perspectives. The authors used a strat-ified random sample of students from Grades 1 to 9 who were assigned to groups based on whether they were from pr imary grades (l-3), middle grades (4-6), or junior high grades (7-9). Ninth-gra de st udents who had experienced the no min al group process acted as facil itators .
The group members were first asked to list the problems that they faced at school on a sheet of paper. Using a round-robin format, every student identified one problem, adding a new item to a list on a flip chart, until it was agreed that the list was complete. From this list, each student identifi ed th e seven most serious problems and rated their severity. The facilita to r then calculated the group ranking of the items.
The 1·a nkings of concerns varied across the d iffe rent age groups. Figh ting and prob-lems with teachers were priority issues for Grades 1 to 3, fighting and drugs were the most serious to Grade 4 to 6 students, and pregnancy, drugs, and drug deals were the strongest concerns for the junior high students.
Teachers from the stu den ts' schools also participated in nominal groups, registerin g additional concerns about stu dent issues such as low parental support, parental probl ems, and lac.k of mot ivation. Both students and teache rs expressed a sense of powerlessness in addressing the pro blems that th ey were identifying. While the nominal group iden tified needs very effectively and in a way that encouraged partnership, it was but the first step in the process of change.
Community Forum Approach
Large open public hearings or community forums may be used to g<1th er information fro m the diverse individuals comprising a community whose needs are being assessed (McKillip, 1998). Similar to a town ha ll meeting lasting for several hours with large numbers of par-ticipants (sometimes 50 or more), this method aims to ensure thaL the b roadest possible sampling of opinions results in a data set reAecting a community consensus regarding the issues being scrutinized. Clearly, this approach aims to give a vo ice to all community memhers, including many who are immediately affected by th e problems of interest.
Witkin and Al tschuld (1995) note tha t special leadership skiJJs are vi tal to the success of this approach. No t eve ryon e has the skills to facilitate large meetings that encourage group members to participate actively and trust that they can openly share ideas that may be different from the majority view.
The advantages of community forums include the fact that they arc a relatively inex-pensive way to hear from large numbers of interested individu als. Another advantage is that pubHc meetings serve to sensitize lhc gen eral public to the problems or to hi ghl igh t potential resistance to proposed solutions. Also, engaging a cross section of community
158 PART I I • QUAN TITATIVt APP ROAC HES: TYPFS or SIUDIES
members may have valuable secondary benefits. For example, when the time co mes to implement recommendations, important people may have bought into the changes bein g suggested.
A pr imary disadvantage to this method is th at there is no means of ensu rin g rhat the participants are a representative sample of their community. lndeed, this will normally not be the case: The ideas and perceptions collected will be those of people who, for some reason, are motivated to infl uence what happens. Citizens who are less interested will not attend and will not be heard, even though they may have reactions to the needs being assessed and th e eventual recommendations for dealing with them.
reeds assessments using a community forum approach were rarely described in the literature. The followi ng study is an in triguing model of usin g the forum t·o both present the results of a previously conducted needs assessment to the research respondents, which subsequently used th e forum to further refine and develop action plans arising from the initial study.
The author included a community fo rum as one aspect of a study of th e relative con-trib ution of social cl ass and ethn icity Lo the differential functioning of Puerto Rican elders in Springfield, Massachusells. The first phase of the study consisted of ind ividual interviews with 591 elders. The purpose of the co mmunity forum was both to present a number of the research respondents with the interview resu lts and to ask subsequent questions. A total of 41 elders parti ci pated in the fo rum.
The forum discussion questions grew from the quantitative survey findings. Two of the five forum questions were, "Why are cultural support systems not playing a mo re active role in helping elders with their needs?" and "Why are churches not playing a more active role?" The author highlighted the extent to which the forum participants assisted in inter-preting the prev ious quantitative needs assessmen t findings, resulting in rccom menda tions that fit closely with this unique and previously marginalized community.
Triangulation of Needs Assessment Information To obviate risks of bias from usin g limited so urces of information, Yegidis, Weinbach, and Morrison-Rodriguez (1999) recommend "triangulatin g" or "collecting data from three (or more) sources, each having different perspectives!' Witkin and Altschuld ( 1995) argue for this step more strongly, suggesting that studi es using only o ne method sho uld be con-sidered seriously flawed .
'lo illustrate, suppose one is resea rching the need in a community for a shelter for women. One could co ndu ct a survey of key informants (social workers, the police, women's organizaLions, self-help group lea ders, etc.), host a community forum, and per form a secondary analysis of existing data in police and social service agency files abo u t the incidence of women requiring shelter. Congruence in the perceptions obtained from each study aspect would clearly represent a stronger case than if o nly one source was accessed. rr th e res ults contradict each other, however, the resea rcher has th e difficult task of assessing which set of perceptions enjoys Lhe grea test credibil ity.
A number of the examples provided in this chapter illustrate the use of more than one m ethod of data collection. An assessment of the health needs of transgendered individu-als in Philadelphia (Kenagy, 2005) is a case in point.
Transgendered individuals have or are in the process of undergoing medical proce-dures to change their gender. Wbile this process entails considerable involvemen t with the medical system, it is not clear what health needs transgendered people experience
C HAPTER 9 • NEEDS A SSESSMENTS 159
afterwards. Kenagy (2005 ) used face- to-face interviews or self administered mail surveys with 182 indivi duals (1 13 male-to -female and 69 female-to -male in dividuals) .
The responses to queries abo ut hea lth needs were p ri marily concerns with respect to HIV/AIDS status {6% were HIV positive), risk ofHIV infection or rein fection (60% had engaged in unprotected sex within the past year), suicide (30% had attempted), their experiences of violence (54% had b een forced to have sex; 52% had been physically ab used), and access to health care (about two thirds had a do cto r or primary care p hysi -cian). Importantly, though, more than one quarter of the informants had been refused medical care because they were transgendcred. The results were used to argue the need for p revention services specific to the transgendered population.
It is n ot uncommon for va rious const ituents to h ave different views about needs. As noted by Revere et al. ( 1996),
It is relatively easy to decide tha t a starving man needs food or a homeless person n eeds sh elter. But what if an assessme nt points to areas of need tha t are n ot acknowledged by the individuals themselves, who may believe they need so mething else altogether? What if the target population and the service providers in the com-munity recognize different areas of need or d isagree as to what will best meet that need? (p. 4)
For example, a needs assessment standardized instrument was developed by M. Carter, Crosby, Geertshuis, and Startup ( 1996) to assess the needs of people with a chronic mental illness. Thiny- two q uestions m eas ured perception s as to whether clie nts needed assistance with a va riety of tasks and issues, such as shopping and cooking, famil y relationships, making use of spare time, and motivation.
Two forms were developed, one for clients and one for staff members, so that the perceptions of the clients and the key informants could be tr iangulated. The results showed poor agreement between workers and clients on many items, suggesting critical differences in perceptions. The issue in such a case is not who to believe but how to address the discrepancies so that they do not negatively affect services.
Ano ther sense in which discrepan cies can b e problematic has lo do wi th w ho priori-tizes needs once they are identified. How and whether research results get used is often a political decision, and different social issues are given importance al different times. For example, child abuse has existed throughout recorded history, but its perceived importance as a problem varies considerably over time and place, and the reso urces avail-ab le to reduce it a nd ameliorate its effects fl uctuate as well.
Over the past several decades, we have come to recognize the surp risingly large number of children who have been sexually abused. Identifying this problem entailed measuring its prevalence and also clarifying Lhe needs of child victims, for child welfare intervention and psych otherapeutic help and so on . As these efforts absorbed scarce research resources, some have argued that o ur push to assist sexual abuse victims has been given such a strong prior-ity that we have not attended sufficiently well to the needs of other mistreated children, those who are negl ec ted or who witness violence between their paren ts, fo r example.
Implementing the Results of Needs Assessments
Although needs assessments are, by d efini tion, research with practical impli cati ons, e nsur ing that the results are imple mented is freque ntly chall enging. Several issues are part
160 PAin II • QuANTITATi v~ API'ROACHES: TvPes Ol· STuDIES
of this general problem. First, the results must be presented in a form in which the sug-gestions and how to implement them are clearly outlined. Th is has been problematic in the past, as C. Carter ( 1996) noted in her review of needs assessments from the late 1970s through 1989. She found that, while authors usuaHy detailed the research process, they offered few s uggestions about how to carry out the required changes. Second, Carter observes that researchers often write for other academics. Important as the academ ic a udience is, with needs assessments, the people we want to i.nlluence with our work include service providers, policy makers, and th e target pop ulation. T hese groups may require a report different from professors and their students. The organization of mater-ial and the style of presenting findings and recommendations should be sensitive lo the likely interests and priorities of nonacademic readers. Researchers might even co nsider writing more than one repo rt, in the in terests of effec tive comm unica tion with diverse audiences.
Amodeo and Gal (1997) recommend another strategy for facilitating utilization of needs assessment research, which is to involve the sponsor organi7.ations in all steps of the study. T his ensures that the sponsor is knowledgeable about the research and committed to fol-lowing it up effectively. In their discussion of this theme, Amodeo and Gal p ropose that researchers should allot a generous amount of time after data collection to help the spon-sor agency digest the findings and plan a response Lo them.
Conclusions
The examples we have offered illustrate that needs assessments are a practical research method with vast potential usefulness to social work. The results can be especially mean-ingful in our work because they document the needs of people experiencing problems, rather than addressing the more esoteric interests of social scientists bent on testing or developing theory. Needs assessme n t ca n raise important questions, identify what aspects of services or policies are useful, indicate what n eeds to be improved, and highlight gaps or misappli ca tions of servi ces. The results ca n challenge us in the same ways that all good J"escarch does, inviting us to test our assumptio ns against evidence gathered from the clients and wmmunities we serve, changing our beliefs and interventions in beneficial ways as a result.
References
Amodeo, M., & Gal, C. (1997) . Strategies for ensuring use of n eeds assessmen t findings: Experiences of a community substa nce abuse preven tion program. journal of Primary Prevention, 18, 227-242.
Brem1an Homiak, K. , & Singleta ry, J. E. (2007) . Famil y violence in congregalions: An exp loralory study of clergy's needs. Social Work and Christia11ity, 31, 18-46.
Carter, C. (1996) . Using and comm unicating findings. In K. Revere, S. Berkowitz, C. Carter, & C. Ferguson (Eds.), Needs assessment: A creative and practical guide for social scientists (pp. 185- 20 l ). WashinglOn, DC: Taylor & Fran cis.
Carter, M., Crosby, C., Ge('rtshuis, S. , & Startup, M. (1996) . Developing reliabili ty in clienL-ccntered me ntal health needs assessment. Journal of Mental Health, 5, 233-243.
CHAPTER 9 • NEEDS ASSESSMENTS 161
Gerdes, K. E., & Benson, R. A. (1 995). Problems of inner-city schoolchildren: Needs assessment by nominal group process. Social Work in Education, 17, 139-147.
Kenagy, G. P. (2005) . Transgender health: Findings from two needs assessment studies in Philadelphia. Health artd Social Work, 30( 1 ), 19- 26.
Maslow, A. ( J 970) . Motivation and personality. New York : llarper & Row. McKillip, J. (1998) . Needs analys is: Process and techniques. fn L. Sickman & D. Rag (Eds.),
Handbook of applied social research methods (pp. 261 284). Thousand Oaks, CA: Sage. Naito-Chan, E., Damron- Rodriguez,}., & Simmons, W. (2004). Identi fying competencies for geri-
atric social work practice. journal of Gerontological Social Work, 43( 4 ), 59- 78. Pan, A., Daley, S., Rivera, L. M., Williams, K., Lingle, D., & Reznik, V. (2006). Understanding the role
of culture in dom estic vio lence: T he Ahimsa Project fo r Safe Families. Journal for Immigrant and Minority Health, 8(1 ), 35-43.
Revere, R., Berkowitz, S., Carter, C., & Ferguson, C. ( l996).lntroduction. Iu R. Revere, S. Berkowit7., C. C. Carter, & C. F. Ferguson (Eds.), Needs assessment: A creative and practical guide for social scie11t-ists (pp. 1-12) . Washington , DC: Taylor & Francis.
Stabb, S. (1995). Needs assessment methodologies. In S. D. Stabb, S. M . H arris, & ). E. Talley (Eds.), Multicultural needs assessment for college and university student populations (pp. 51-115 ) . Spri11gfield, IL: Charles C Thomas.
' tracy, E. M., Green, R. K., & Bremseth, M.D. {1993 ). Meeting the environmental needs of abused and neglected children: implications from a statewide survey of supportive services. Social Work Research & Abstracts, 29(2), 21 – 26.
Tutty, L., & Rothery, M. (2002) . How well do emergency s helters assist women an d their children? In L. TuLly & C. Goard (Eds.), Reclaiming self Issues and resources for women abused by intimate partners (pp. 25-42). Halifax, NS: Fernwood and Resolve.
Tutty, L M., McNichol, K. , & Christensen,}. (2008) . Calgary's Homefront specialized domestic vio-lence co urt. In J. Ursel, L. Tutty, & }. LeMaistrc (Eds.), What's law got to do with it? The la"~ spe-cialized courts and domestic violence in Canada. Toron to, ON: Cor morant Press.
Tutty, L. M ., Weaver, G., & Rothery, M. A. ( 1999). Residents' views of the efficacy of shelter services for abused women. Violence Against Women, 5, 869- 925.
Wancata, }., Krautgartncr, M., Berner,}., Scumaci, S., Freid!, M., Alexandrowic7., It, et al. (2006). The "Carer's r eeds Assessment for Schizophrenia": An instrument to assess the needs of relatiYes caring for sch izophrenia patients. Social Psychiatry & Psychiatric Epidemiology, 41, 221-229.
Weave r, H. W. ( 1997). The ch allenges of research in Native American commun it ies: Incorporating princip les of cultural compete nce. Journal of Social Service Research, 23, l – 15.
Wilkin, B. R., & Al tschuld, 1. W. ( 1995). Planning and conducting needs assessments: A practical guide. Thousand Oaks, CA: Sage.
Yegidis, B. 1.., Weinbach, R. W., & Morrison -Rodriguez, B. (1999). Research melhodsfor social work-ers (3rd eel.). Boston: All yn & Bacon.
http:/ /www.extension.iastate.edu/communitics/tools/assess/ This Web site by the Iowa State University Extension Division provid es a primer on five types of needs assessments: existing data sources, attitude survey approaches, key informant approaches, com munity forums, and focus groups.
http:/ /ctb.k u.edu/tools/scctio n_l 042.htm The Com munity Tool Box is a se rvice of the Work Groups for Comm unity Health and Developm ent at the University of Kansas. Th is compre hensive Web site prese nts information on numerous evaluation methods for use in the community, with needs assessments be ing only one.
162 PART 11 • Q uANTITATIVE A PPROACH ES: TYPES Of STUD I ES
DISCUSSION QUESTIONS –
1. Whal are the disadvantages and advantages of quantitative as compared to qualitative needs assessment methods?
1. What are the advantages and disadvantages of asking key informants about the needs of an iden-tified population versus speaking to those affected by the problem them selves?
3. Compare community forums and focus groups. For what pu rposes might one be preferred over the other?
4. Discuss the concepts of "needs" and "wants." How can those developing needs assessment ensure that the needs that are bei ng identified are necessa ry and important?
5. What are the best strategies to ensure that needs assessment recommendations are implemented?
