View source for Needs assessment guide


= Problem Definition and Scope =
==What is the problem?==
===Describe context of the problem===
This places the issue that your client group is facing in a larger social and historical context. Purpose is to help indicate the larger significance of the issue.
Eg, vulnerable seniors facing end-of-life issues are just one group out of many others—disadvantaged dying (disabled, poor, mental health problems etc) 

===Discuss existing definitions of the issue to arrive at your own definition===
Sometimes there may be a lack of clarity on the problem area and you may need to discuss how academics, agencies, practitioners define the problem and arrive at a definition that works for your purposes.
Eg At risk youth
Eg End of life care; palliative care

===Identify who your target clients are===
The target client can be defined broadly and in a more encompassing way (youth-at-risk; seniors with mental health issues), or more narrowly (low-income seniors diagnosed with non-cancer terminal illnesses).

===Set inclusion and exclusion criteria===
It is important to set up your inclusion criteria for the target client so that we can be precise about what we mean when we refer to this group. What criteria should they satisfy for them to qualify for your attention? 

This is important to determine how large the group is, and to establish whether existing services are getting to them. To put it another way, how would you know if a person should be part of the target client group when you see one?

===Determine the characteristic & ‘strength’ of your inclusion criteria: what of these criteria are necessary, sufficient?===
Eg, we may want to take care of ‘vulnerable seniors’ and have established a few criteria: but what if they were not poor, but did not have family support? What if they have a maid, but had limited family support? In this case, you may decide that if they satisfy some of the criteria (eg 2 out of 3, or at least one out of all of them etc), they will be included.

If possible, define client group with greater specificity, eg elderly aged 60-75, smokers aged 18-29.

===Describe the typical client’s experience===
It is useful to give an account of the client in a narrative or story-like form, so as to capture and convey the meaningfulness and experience of what a typical client goes through in their everyday life.  [We will leave it to later for a more quantitative breakdown of demographic characteristics]

===Conceptualize the desired outcomes for these clients===
Don't define the problem as a need for a programme or service. Do not merely define the problem in terms of insufficiencies with the existing services, eg not enough childcare services, lack of integration across services, low capacity, lack of skilled manpower etc.

Instead, be clear about ultimate outcome you want to achieve. What outcomes are desirable for these clients? If you start with outcomes first, so that you remain open to different ways of getting there; consider alternative programmes, interventions, means to achieve stated outcomes.

==Size and significance of the problem==
First, you can start by searching for statistics from various government sources and document them in your catalogue first. Once you have everything in place, you can analyze the information. What you are trying to get at is a sense of how big the problem is.

Total number of target clientele when available OR (prevalence rate x population)à size of the universe (size of total potential need/demand for services). This conveys the size of all those who are eligible for intervention

===Prevalence (existing size of problem)===
Prevalence-number of existing cases in specific area during a specified period of time
[Expressed as a percentage, eg, the prevalence of drop-outs from secondary schools in Singapore is 2%]

You may need to apply differential prevalence rates for segments of the community – e.g. frail elderly have higher rate of mental illness than young elderly (Kettner et al 2008: 75)

===Incidence (new cases)===
Incidence-number of new cases in specific area during a specified period of time (usually a year). This tells us the recent additions of population in need. The incidence rate is the number of new cases divided by the size of the population under consideration.
[eg in 2012, if there were 20 new cases of school drop-outs from secondary schools in Singapore, divide that by the number of secondary school students in Singapore (let’s say 200,000). The incidence rate would be 0.0001 or 1 out of every 10,000 ]

===Size of expressed need / demand===
The utilization rates of specific services can be used to estimate need and demand. The utilization rate and the service rejection rate is an indication of the size of expressed need, that is, demand from those willing to seek out services. However, this is only an approximate indication of the size of the actual demand, because those who have needs may not actually seek help for various reasons (eg pregnant teenagers who want to escape stigma; or a service dominated by clients from a different ethnic group). Therefore, it may not be a good estimation of actual unmet need.

These numbers are only calculated for the purpose of conveying a general sense of the size of the problem, but the more detailed and specific calculations for each specific category of need can come later

===Consequence and significance===
Spell out the short to long term consequences of the problem. If you keep asking the ‘so what?’ question you may be able to articulate longer term consequences of the problem that will communicate the importance and weight of the issue. Eg elderly not aging well, but so what? Answer: health care costs go up.

==How is problem likely to change?==
Social indicators, trends or projections...Is the total unmet need rising, slowing or stable?

===Trends [How has the problem been changing in previous years?]===
Track change from previous years.
Sometimes funders will also ask what the rate of change is.

===Projections [How will the problem change?]===
See Synthetic Estimations (McKillip 1998)

====Projections using Historical Data====
-time series data: positive, negative, seasonal trends, cyclical trends
-adjustments done to smooth out fluctuations (eg seasonal fluctuations)
-method of least squares (regression)

====Projections when history is inadequate====
-Single Factor Projections
-Judgmental Methods: eg Delphi Technique

=Client Profile and Client System=
Function of understanding client profile: To determine what segments or subgroups exist in the overall population, understand the distinctive ways they behave, so as to develop customized services that would serve their unique needs.

==Demographic Characteristics==
To understand the client profile, we can start with collecting demographic characteristics, which are facts about the makeup of a population, such as age, gender, income level, race, ethnicity, religion, occupation, etc. 

Describe the common traits of the client population who are at-risk or already experiencing problem (age, sex, race, marital status, class, housing type).

==Client Segmentation==
Clear ‘market segments’ are made up of client groups with similar characteristics and that have similar needs. Because they are internally homogenous and respond similarly to external stimulus, they become the basis for planning and developing improvements to service. What services (products) would appeal to what client groups (market segment)?

First of all, ask yourself: Are your clients internally homogenous? Are there meaningful and identifiably different segments? That is, are they similar enough to one another such that the services provided to them are equally appropriate? Or are there systematic differences between them, such that services used for some segments would not be fully appropriate for others? 

You do not need to segmentize your clients for the sake of it; you segmentize when you have grounds to believe that a one-size-fits-all solution is not working well. 

==Sketch a portrait of a typical member of each segment==
Sketch a portrait or characterize a typical member of each segment, so as to communicate the subjective meanings and lived experience of these members.
[e.g. types of sex workers: crack whores, street walkers, high class social escorts, undergrads doing it for pocket money]

This need not be extensive and detailed, but enough to capture the circumstances they face, and the motivations and rationale for their behaviour.

==Client System and Ecology==
Function: Understanding social system and ecology that a client functions within, so as to understand the influences and therefore the strategic points of intervention.

===Who are those at risk or vulnerable to the problem?===
If we can identify accurately who are those at risk, we can plan to utilize more preventive measures.

===Who are those indirectly affected, or can contribute to the resolution of the problem?===
Client system: those who are indirectly affected, and/or can contribute to the resolution of the problem (eg family, caregivers, professionals, volunteers, peers). 

We can more comprehensively understand the needs of ‘secondary’ clients that may also need help. Do family members, caregivers and significant others have needs related to problem? If we are clear about primary client group and secondary (eg clients vs caregivers), we can be clear so that you are clear who your programme is supposed to help, and to understand what is your central programme and what is peripheral supportive programmes.

More importantly, if we can identify the client system, we can identify where and who to intervene in. E.g., if we know that for younger children, the family unit has more influence, then we can decide to focus on parenting support programmes. If we know that for youth, peers have more influence, then we would focus on their friends. Therefore, it is important to describe the system of support of the clients—we may need to help the caregivers to help the clients.

===How is the problem distributed in the community?===
-Where are the problems and solutions distributed? Where are concentrations of the problem and target clientele? Where are the services located? 
-Understanding this will help to avoid over-provision (overlaps and duplication) or under-provision (lack ease of access). This will help to better coordinate services and improve integration across services and agencies. 

Investigate further: Quantitative Methods for Spatial Analysis
Methods of Spatial Analysis for Identifying Concentrations of High Risk Groups (Kettner et al 2008)
Overall national average may be low compared to other countries, but in subareas, the indicators may be high. Use of factor analysis (reduce large number of variables into a smaller number of constructs or indicators). If large number of variables are intercorrelated, then these interrelationships may be due to underlying factors.

=Needs Analysis=

==Identifying and Defining Needs==
Approach: Top-down first, then bottom-up
Needs are comprehensively identified through a mix of top-down and ground-up approaches. Environmental scans and trend analysis can give a broad overview of general needs. However, such needs may be removed from actual problems and issues that clients and practitioners face. Therefore, it is important to capture perceived and felt needs from the people directly involved.

1.	Environmental Scanning, Service Utilization Statistics & Literature Review: to find out the range, types & nature of needs [top-down]

2.	Needs Assessment Research:  to find out expressed and felt needs from client & community [bottom-up]

Environmental scans and trend analysis can give a broad overview of general needs. General statistics, prevalence rate, normative & comparative assessments are top-down approaches. This is complemented by bottom-up approaches like the use of service statistics (utilization rate, waitlists) as a measurement of expressed needs. However, such needs may still be removed from actual problems and issues that clients and practitioners face. Therefore, it is important to capture perceived and felt needs from the people directly involved.

But even the use of these bottom-up approaches by themselves will not give a comprehensive picture as we cannot assume that all persons in need will seek help. For example, service utilisation statistics does not provide the true indication of need as some clients might be facing constraints to utilisation (e.g. no transport to access Senior Activity Centres). It is therefore important to reach out to others beyond those who use the services, to get a sense of needs from the ground up.

==Community Needs Assessment==
Locality or neighborhood based

FSCs and CDCs are best placed to do this: measure and track the full range of needs of the diverse residents within their service boundaries.

A way to determine community needs by proxy is to derive them from examining the services provided in one area to one population and using this information as the basis to determine the sort of services required in another area with a similar population.

However, we have to be careful because populations which are similar demographically can have different latent and expressed needs. 

As a result, proper data collection should complement such information by setting out to capture felt community needs directly from the community.

==Client Type Needs Assessment==

==Categorizing Needs==
This is to help organize and analyse the main areas of needs so that we can make sense of the diversity of opinions and studies that unearth a wide variety of needs that clients have. Categorizing such needs would be useful when utilizing the Resource Matrix (see Guide 3) to assess how well existing communal resources are serving these needs.

Taxonomies of Social Service Programmes & Needs:
See United Way: A Taxonomy of Social Goals and Human Service Programs.
http://www.211taxonomy.org/resources/library

This is a taxonomy of social goals, the human service system, and the type of programs within those systems, that taken together as a whole, are designed to achieve those broader goals. 
Eg social goal=optimal social functioning
System=individual and family life services; social adjustment and social development; cultural and spiritual enrichment etc
Service=Family preservation and Strengthening; Family Substitute (Foster Care); Family Supplement
Programs=Counselling, Single-Parent Family Development

There is a program index for target groups (see Appendix D, page 289) for a list of programs exclusively designed to assist a target group (Aged, Children and Youth, Handicapped, Low Income, Offenders etc).

===Possible categories===
Public awareness & acceptance

Screening, Detection & Diagnosis
-proper detection, assessment & diagnosis

Information & referral
Do target clients (and caregivers) have understandable info to guide decision making & planning for their problem or area of need?

‘Treatment’
-Psychosocial needs
-ADL
-Education
-Employment [sheltered workshop, JPJS]

Integration of various treatments and of flow-through

‘Aftercare’

Capability Support
Research & Advocacy

Professional Competence

Organizational Development

Caregiver or Volunteer Support
-knowledge & skills training
-respite
- social support

Misc
-security: financial, housing etc
-transportation and mobility
-legal

=Prioritization=
How do we determine the importance and significance of the problem once we have identified and defined it?
Purpose: To develop criteria and heuristics that can be used to prioritize the list of needs and gaps to assist in resource allocation decisions that need to be publicly accountable to our multiple stakeholders.

Needs analysis also involves evaluating and ranking needs (Alston & Bowles 2003: 124). In the previous section, we have identified different needs, but we have not yet put all these needs onto a single conceptual plane to determine what is more or less important. 

This framework is not an algorithm that will systematically transform identified needs into a prioritized list. Such an algorithm is difficult because prioritization of human needs depends on a kind of judgment that is informed by normative values that are diverse and not universally shared.

As scholars note, this is often a political act, and therefore, a political act requires consensus building rather than scientific determination.

===Possible criteria===
•	Frequency (occurs often)

•	Duration (been around, is persistent)

•	Scope (affects a lot of people, proportion affected)

•	Severity (degree of suffering and pain)

•	Urgency (time sensitive)

•	Perception (whether people think it is a problem; a national priority?)

===Comparative need and normative need===
Comparative need
Purpose: to determine the significance of the need according to normative standards and in comparison to other communities and contexts.

Comparative need is measured by reference to a user already receiving the service in question. Therefore, a person is in comparative need if he or she has the same or worse characteristics as someone receiving the service. The concept also can be applied to districts (for example, district A provides free medical treatment while district B does not) or to countries. However, this method of comparison leaves two questions unanswered as only existing services are being compared. “What if there is a need for a new service?” “Does it also imply that the reference standard is faultless and no longer needs improvement?”

Nature, size and trends of needs should be assessed according to compared to other similar communities to determine how significant it is. [Similarly, for determining the adequacy of solutions in addressing need (see Guide 2), it is also important to compare the access and quality of services in other similar communities]

Comparative need is derived from examining the services provided in one area to one population and using this information as the basis to determine the sort of services required in another area with a similar population. When assessing comparative need the level of service provision in the reference area must be appropriate in the first place. Be cautious that data collected may in fact be due to over-servicing or under-servicing by service providersrather than an indication of true need for the service by health consumers.

Social indicator analysis: to compare regions or communities with others to determine area variations and target group variations (Alston & Bowles 2003: 131); historical trends; size of potential service users; population characteristics. Helps outline problem, but cannot help devise solutions.
Social Indicators, social trends or demographics of the community (e.g. age, gender, family size, employment status). (e.g. crime rate, school drop-out rate).
Normative need
A normative need is established by custom, authority or general consensus eg estimate rate of serious violence in families by use rate of incidence per population. Identified according to a norm (or set standard); such norms are generally set by experts and usually determined according to some criterion (e.g. standards of unfitness in houses).
Normative need tends to be professionally defined and has a knowledge base. A desirable standard is set by professionals, policy makers or social scientists, against which the actual standard is compared. Those below the standard are said to be in need of support and special services. A good example is the intelligent quotient (IQ) which is used to indicate people with special needs (below a score of 80 is defined as moderately retarded). Social security entitlement is also normatively defined. People’s need is measured against their assets. Only if the asset value is below a set amount, which is defined by policy makers, then eligibility results. Indeed the setting of the amount is not value-free, it is relative to the socio-political and economic situations, and may change from time to time. 
In evaluating societal established standards, if the community is at or above those standards, then there is no need (see Kettner et al 2008: 52).
Assessment of Normative Need by Extrapolating from Existing Studies. Using prevalence rates to estimate. But need to apply differential rates for segments of the community, eg frail elderly have higher rate of mental illness than young elderly. Also, be aware of the definition of the concept, ‘mental illness’ may be defined differently for different studies. Also possible to complement this by inviting expert (who will be familiar with statistics & research) to propose specific strategies and suggest reasonable levels of service provision (Kettner 2008: 75).
Normative Need – Identified according to a norm (or set standard); such norms are generally set by experts and usually determined according to some criterion (e.g. standards of unfitness in houses). (Bradshaw, 1972)

===Possible principles of prioritization===
====Principle of prioritization #1: basic needs first====
Deal with basic needs first before dealing with esteem and belonging (derived from Maslow). For example, family violence programs (see Kettner et al 2008: 53).
-family violence programs: deal with basic needs first before deal with esteem and belonging (derived from Maslow) (see Kettner et al 2008: 53)
-Rather than spending resources on developing sophisticated technologies,  priority is to primary medical care and health services for the general population (philosophy derived from Ponsioen’s 1962 notion of need) (see Kettner et al 2008: 53)

====Principle of prioritization #2: general need more important than specific need====
For example, rather than spending resources on developing sophisticated technologies,  priority is to primary medical care and health services for the general population (philosophy derived from Ponsioen’s 1962 notion of need) (see Kettner et al 2008: 53)

==Heuristics for Decision Making==

===Heuristic 1: ranking===
To help groups reach consensus on the priorities of needs, we can utilize ranking:
“Ranking is a process that encourages individual stakeholders to develop their own ranked priority list of need problem areas for the target population. Individual listings of ranked needs are then summed to create a composite or summary ranking” (Petersen & Alexander 2001: 78).

However, the fundamental weakness of the ranking methodology for establishing need priorities is that it pits the needs of specific special interest groups against each other...One means of addressing this issue is to segment, or subdivide, target populations or general need areas prior to ranking to ensure that the needs of specific target population subgroups, eg infants, teens, adults, women, elderly, and children with special health care needs, are addressed. Need rankings can then be developed for each subpopulation (Petersen & Alexander 2001: 79-80).

Precise, but difficult to determine criteria and weightage for measurement. In all likelihood, precise ranking of needs would be unnecessary since what is actually needed is a broad sense of the amount of time and resources to be invested in a few key areas. Knowing the fine-grained details of how specifically one need ranks higher than another will not drastically affect the amount of resources invested in them if we know they are important problems that both need to be addressed.

Individual stakeholders develop own ranked priorities and individual listings of ranked needs are summed to create composite or summary (Petersen & Alexander, Needs Assessment in Public Health, 78).

However, the weakness of the ranking methodology for establishing need priorities is that it pits the needs of special interest groups against each other (Petersen & Alexander, Needs Assessment in Public Health, 81).

Criteria for ranking could consider:
•	Size of the problem
•	Seriousness of the problem, etc

===Heuristic 2: Alignment between different information on needs===
Real need is determined when normative, felt, expressed, comparative need is present.

Bradshaw (1972) provided a methodology in making a ‘real’ need possible. His proposal was to first delineate four types of social needs, namely, (a) normative; (b) felt; (c) expressed; and (d) comparative, then to examine their presence in a given situation. The presence of all types of needs is equated to real need. He proposed a taxonomy of need in which the four need-types, when considered in a reality need situation, were each assigned a plus sign (presence of need) or a minus sign (absence of need). Real need is defined as presence in all four standards. The taxonomy gives rise to 12 possible combinations (for example, ++++, ----, ++--) which helps in decision-making. 

Bradshaw’s approach to need is a useful framework for policy-making and for analysing policy to the extent that political, economic and social factors can be taken into account in deciding needs and services.

===Heuristic 3: Indicate how many principles a programme satisfies===
A practical but less precise way for decision making is to include in our needs & gaps report a checklist of all the criteria used to assess the importance of a need. For each need, we would indicate how many of the above criteria have been satisfied, without necessarily making a case that one need is more important than another 

===Heuristic 4: Effort-Impact Analysis===

===Heuristic 5: Decision-Matrix or Optimal Criteria Matrix===

=Types of Information Required from Various Stakeholders=
Government Agencies
VWOs
Clients
etc