Evaluation of the content of assessment frameworks used for screening prospective foster parents in the north west province, South Africa

Purpose of the study: To propose a universal assessment and screening framework for prospective foster carers within Dr. Kenneth Kaunda and Ruth Mompati districts in the North-West Province, South Africa. Methodology: This study used a quantitative, descriptive, and exploratory approach, principally directed by an evaluation paradigm. The above-mentioned strategy was backed by the adoption of a cross-sectional survey design (Creswell, 2014). The researchers were able to quantify the degree of agreement between two or more respondents thanks to the fully-crossed design. To examine the inter-rater agreement of the five specified assessment frameworks, data were collected from respondents at one point in time using a self-developed, paper-based survey using a 5-point Likert scale. Main Findings: As seen by differences in sample frameworks according to the quantitative component of this study, there is a lack of standardization in the use of assessment frameworks during the screening and assessment of potential foster parents. Furthermore, these frameworks were not always subjected to scientific scrutiny and best practice evaluation. Applications of the study: There is a need to standardize processes that will lead to higher-quality practices and client services in the long run. This research will help to develop a consistent, uniform baseline framework for assessing foster placements, as well as identify viable screening instruments and improve the structure of assessment processes. Novelty/Originality of the study: This research proposes that by identifying ideal framework elements and gathering information regarding them in existing assessment frameworks, evidence-based practice in foster care, specifically in the screening and assessment of potential foster parents, can be improved. descriptive various assessment frameworks, as the three recommended for


INTRODUCTION
Foster care appears to be the greatest alternative for children who need to be taken care of outside of their homes all over the world. When a child is removed from the care of his or her biological parents, it is critical to ensure that the child is placed in conditions that are at least better than those from which they came (Casey Family Programs, 2018). Prospective foster parents must be thoroughly screened and assessed in order to determine their appropriateness and compatibility (Carter, 2013). Given that the number of children in need of alternative care in South Africa continues to rise, social workers are frequently unable to devote enough time to adequate screening and assessment of prospective foster parents, resulting in large caseloads (Breen, 2015;Murray, 2016). These factors add to the burden on social workers to manage ever-increasing caseloads while still making educated decisions about a high-quality and accurate placement in the best interests of the child. To keep up with changing caseloads and the rising demand for services, social workers rely mostly on their own personal experience and practice wisdom (Holland, 2011). Alper and Howe (2017) point out that, placement accuracy and "fit" are less predictable and scientific in some cases, and more of an art in others.
According to our observations, placement decisions are frequently based on the observational skills and wisdom of a social worker who has conducted these assessments for years and is less concerned with the scientific foundation that should support such a procedure. We don't argue against intuitive knowledge or gaining a sense of caregiver suitability, but we do advocate for a combined scientific and artistic effort. We also understand the necessity of conducting foster parent screening in the context of professional human interactions and that evaluation should not devolve into a boxticking exercise for the sake of creating a speedy court report. However, with the current increase in case overload and pressurized atmosphere, we doubt that the relational analysis envisaged in the in-depth relational context is achievable. A preliminary investigation into the types of instruments and practices used by welfare organizations in screening social workers revealed that they frequently used self-developed and, by extension, non-standardised assessment frameworks with inconsistent content, procedures, and implementation across different practice contexts (Murray, 2016). Multiple factors, some of which may be the motivating factor for the prospective foster parent and their characteristics, concerns about the child's anticipated behavior problem, the attitude of social workers dealing with prospective foster parents, and the foster parent's parental skills, should not be underestimated when it comes to placement outcome (Ntshongwana & Tanga, 2018). These researchers concluded that screening and assessment should begin during recruitment activities aimed at potential foster parents, as this is when the relationship-building process begins. This procedure with potential caregivers plays a critical part in the screening and evaluation of potential carers, demonstrating that the organization's screening and assessment method is not the sole element.
However, from the standpoint of professional practice, we believe that the lack of consistency in screening and assessing foster parent appropriateness contributes to irregular reporting methods in children's court cases, which can lead to improper placements and likely placement failures. Because screening and assessment processes are significantly linked to effective foster care placement, they should not be left to the discretion of a single social worker (Pollack, 2012). Using various sources of information and applying precisely specified assessment frameworks, social workers may complete regular evaluations such as child foster care placement possibilities in a systematic and standardised manner (Luke & Shebba, 2013;Murray, 2016). As a result, using systematic assessment frameworks can provide consistency in data collection processes that are both informative and useful for making decisions about the fit between a child in need of care and a foster parent's capacity to accept and work with the child in a foster family arrangement. A methodical framework can also ensure the testing of hypotheses and the acquisition of information about the functioning of the potential foster parent that would otherwise be unknown to the practitioner (Holland, 2011). Most importantly, assessment frameworks assist in predicting the fit in the future post-placement and whether the placement has the potential to become long-term and permanent, whether the child has potential special needs arising from developmental trauma that require extraordinary effort, skill, or intervention capacity that should be managed effectively during placement, and whether the child has potential special needs arising from developmental trauma that require extraordinary effort, skill, or intervention capacity that should be managed effectively during placement (Luke & Shebba, 2013).

Foster care in South Africa
South Africa is characterized by a wide collection of conditions, including poverty, inequality, child abuse, domestic violence, and unemployment, all of which lead to children ending up in foster care. According to statistics, there are about 19.6 million children in South Africa. Nearly 2.8 million of these are orphans, accounting for 14 percent of South African children (University of Cape Town, Children's Institute, 2019). In 2016, there were an estimated 521 055 children in foster care (Breen, 2015). Given the figures above, it is apparent that South Africa's child care system is under severe strain as the number of children in need of alternative care continues to rise (Breen, 2015). Many social workers who screen and assess potential foster parents had only in-service training from their supervisors, who in turn often have minimal specialized training in screening and assessing foster parents (Carter, 2013). During the screening and evaluation process, the social worker's in most instances collect data guided by practice wisdom, which is based on replicating and institutionalizing current working practices, and is frequently used in decision-making (Holland, 2011).

The role of screening, assessment, and assessment frameworks in foster care
The screening and evaluation of potential foster parents is an intense information-gathering procedure marked by rigor and a strong scientific component, intending to evaluate if the prospective foster parents can care for a child who requires alternative care (Carter, 2013). The use of standardised assessment tools to assist professional judgments is important, according to Hepworth et al. (2013), since it accounts for human error, individual biases, personal histories, values, and behaviors. According to Ntshongwana and Tanga (2018), screening methods should evaluate coping strategies, behaviors, attitudes, sentiments, and beliefs in a variety of domains, as well as a person's financial situation and social network health situations. As a result, in order to connect the many threads of the prospective foster parent screening and evaluation process, standardization and uniformity of the procedure are critical, as they are more likely to address the issue of probable inconsistency in practice (Murray, 2016). This circumstance necessitates the development and implementation of particular evaluation frameworks to improve the quality of prospective foster parent screening and assessment.

The difference between screening and assessment
In foster care, the phrases screening and assessment are commonly used interchangeably, however, they serve a rather different purpose. Screening, on the one hand, is viewed as a quick, focused investigation in which information about the child's functional characteristics is gathered and compared to information gathered from prospective foster parents to determine if (1) the prospective foster parents meet the stated requirements and (2) to identify critical areas for a comprehensive assessment (The Children's Bureau & the Department of Health and Human Services, 2014). In this case, screening can help the social worker identify the sorts of circumstances that are likely to trigger the child's primary care traumatic experiences and then communicate that knowledge with the potential foster parent so that they are completely aware of the child's needs. In-depth knowledge about the intrinsic functioning of prospective foster parents is also important. Therefore, making a screening process the first step before an evaluation.
Whereas assessment, on the other hand is a data-gathering process that occurs between a social worker and a client, in which the data is analyzed for significance and synthesised into usable recommendations or judgments about the parties' future (Hepworth et al., 2013). The developmental requirements of a child, the home environment, and any other conditions that may have an impact on the child's need for protection and therapeutic services" are examined in this procedure (Children's Act, 2005). As a result, the assessment process focuses on two primary stages, among others. First, acquiring essential information on potential foster parents in order to determine their fitness to care for the child; second, getting thorough information about prospective foster parents and their circumstances is likely to have an impact on their overall functioning. The social worker can examine the child's potential 'fit' and the prospective foster parent using these two key areas. A strong placement fit can lead to placement stability, whereas a weak placement fit can lead to placement disruption or even breakdown.

The importance of utilising screening and assessment in conjunction with assessment frameworks
Data-based practice is defined by Drisko and Grady (2015) as a method in which social workers use the best available research evidence with their professional experience, client qualities, values, preferences, and circumstances to make informed practice decisions. Therefore, standardised assessment can strengthen the evidence-based practice when used during the screening and assessment process of the child in need of care and the prospective foster parent. In addition, Adams et al. (2014) define standardised assessment as a method of acquiring information about persons that is consistent across subjects, locations, and administrations, essentially reducing assessment inconsistency. Thus, using standardised assessment procedures assist in reducing exceptions in case management and reaching conclusions that allow comparison across populations and the development of norms that can serve as benchmarks in future cases.
Furthermore, measures used in the standardised assessment, known as assessment tools or instruments (assessment frameworks for this study), consist of a collection of procedures followed during the assessment that spans several dimensions or critical points in need of address during the assessment. Assessment implies that measurement instruments can enhance the uniformity and standardisation of the assessment process, and if captured in an assessment framework, contributes to uniform assessment and the creation of a basis for comparison. Social workers can use evaluation frameworks to help them identify and assess potential foster parents. Assessment frameworks' ultimate goal is to forecast the long-term viability of child placements (Luke & Shebba, 2013). As a result, an acceptable evaluation is a multifaceted process including several sources of information, and if systematic techniques are utilized in such assessments, they represent a high degree of standardization and comparability (Hepworth et al., 2013). Thus, any screening or assessment must gather information from a variety of sources, which includes but is not limited to interviews, direct observation, tests or assessment frameworks, and the social worker's personal experiences based on direct interaction with successful foster parents to conclude (Murray, 2016).
Therefore, it is of considerable importance that the social workers utilising any assessment framework must have the necessary knowledge regarding test theory and the characteristics of the assessment frameworks they use (Hepworth et al., 2013). The reason is that many assessment frameworks show biases, have poor reliability and validity, and are unsuitable for specific contexts or populations (Hepworth et al., 2013). As a result, social workers who utilize any assessment framework must have a thorough understanding of test theory and the features of the assessment frameworks they employ (Hepworth et al., 2013). The reason for this is that many evaluation frameworks include biases, are unsuited for certain contexts or groups, and have low reliability and validity (Hepworth et al., 2013). Most organizations, for example, utilize non-standardised metrics such as questionnaires throughout the screening and evaluation process. Assessors want potential foster parents to respond to specific questions on various issues, such as whether they believe they will offer excellent care for the foster child. What are their strategies for dealing with family stress? What kind of problem-solving abilities do they have? Assessment frameworks can measure potential foster parents' nurturing ability rather than depending on their perhaps biased answers (Holland, 2011).
As a result, research has revealed that, despite the numerous benefits of using evaluation frameworks, few social workers opt to do so (Arbeiter & Toros, 2017;Placido & Cecil, 2014). Furthermore, according to the findings of a research done by Khoza (2011), the issues impeding the application of evaluation frameworks in the intervention are attributable to macro and micro-level restrictions. On the one hand, macro-level variables included departmental bureaucracy, inaccessibility of assessment frameworks, and a lack of ability and resources to aid the process, all of which, according to the study, might contribute to the lack of uniformity in assessment services (Khoza, 2011). Furthermore, micro-level problems included a lack of supervision, social worker burnout, a lack of skills and knowledge, and, finally, a lack of and absence of strategies for implementing the assessment framework (Khoza, 2011). These reasons reinforce the claim that social workers are frequently forced to rely on practical expertise and experience to assist them in this respect. Lastly, these characteristics highlight the lack of consistency among practice contexts, specifically in terms of the types of evaluation frameworks used, if any, and how the content of these frameworks compares to one another.
Given the ever-changing nature of the South African welfare environment, it's apparent that these developments have a significant influence on the child protection system. As a result of these developments, social work practices must adopt a new approach to children in need of foster care settings. Macro-level change can provide an enabling climate for employing evaluation frameworks at micro-intervention levels. This environment can be impacted by coordinating the evaluation frameworks utilized by various governmental and non-governmental organizations in order to achieve uniformity (Khoza, 2011;Toros, 2017).

The basic requirements for an assessment framework
The physical, emotional, psychological, and spiritual domains are the most prevalent areas of well-being assessed during screening and evaluation (Zastrow & Kirst-Ashman, 2013). These categories also assess foster parents' abilities to give unconditional love, protection, and security, as well as stability and nurturing (Carter & Van Breda, 2015). The study hypothesizes that these domains fluctuate depending on the practice situation. Foster families confront distinct obstacles in today's culture, necessitating a specific assessment approach. Unfortunately, there is little research that has revealed metrics with strong psychometric qualities that may be used to assess foster families (Adams et al., 2014; Skillmark & Oscarsson, 2018; UNICEF & UNHCR, 2013). However, there is a worry that these efforts are insufficient to address many of the particular issues that the present foster care system faces (Cuddeback et al., 2007). Alper and Howe (2017, p. 51) presented a range of variables for evaluation, including parental goals and expectations, sensitivity and warmth, stress and its management, and the composition and quality of family support networks.
Functional criteria, technical requirements, and usability requirements must all be met in order to employ an assessment framework effectively in the foster care setting. Each of these sections provides research findings on the characteristics of an effective assessment framework. The following prerequisites are numbered only to make it easier to read them. As a result, the order of choice in these sectors is not shown by the numbers.

Functional requirements
(1) Most importantly, build a fundamental model of what characterizes "good" and "bad" foster parents (often referred to as behavioural profiles) ahead of time (Hepworth et al., 2013;Carter & Van Breda, 2015). Precise characteristics must be in place since they will aid in the prediction of more successful placement outcomes. (Luke & Shebba, 2013). (2) The assessment framework's scope and emphasis must be exact. The scope and focus of the social worker's work may differ based on the social worker's function, the location in which he or she works, and the client's requirements (Hepworth et al., 2013). (3) The assessment framework must be culturally appropriate (Hepworth et al., 2013;Khoza, 2011). (4) The framework must collect data from a variety of sources in order to allow for triangulation (Holland, 2011;Hepworth et al., 2013). (5) when it has both quantitative and qualitative elements (Maruster & Gijsenberg, 2012). (6) The quantitative and qualitative elements must complement each other; some questions are a mix of open-ended and closedended questions. Foster candidates must be able to develop their own replies from a set of options by allowing openended questions (i.e. "other" or "comments" must be an option). The benefit of this form of the question is that the respondent can still record a suitable response if the alternatives provided are inadequate (Stats NZ, 2019). (7) The assessment framework must have a strong theoretical foundation and must employ one or more theoretical orientations to guide the evaluation, such as solution-focused treatment, cognitive behavior therapy, or systems theory (

Technical requirements
The following are the technical criteria of an assessment framework: (1) The evaluation framework has excellent psychometric features (Borsboom & Molenaar, 2015;Russel,2015). (2) Validity of the face and content must be shown (Hepworth et al., 2013;Simula, 2016). (3) The assessment framework's dependability must be proved (Rubin & Babbie, 2013). (4) The prospective foster parents are provided a guideline with defined performance requirements that they must satisfy (Australian Government, 2015). (5) Prospective foster parents must provide performance proof by speaking, performing, or writing anything that demonstrates their expertise (Australian Government, 2015). (6) Finally, the assessment framework must employ explicit assessment decision-making principles to determine if potential foster parents are qualified for the job (Australian Government, 2015).

Usability requirements
Users will be able to access the assessment framework's functionalities and successfully use it if it is usable.  (2012) state that the assessment framework must be adaptable to varied practice scenarios. (7) Finally, the evaluation system must be effective in order to achieve high productivity (Simões-Marques & Nunes, 2012). A framework must be simple to use, and it must allow for the collection of acceptable and relevant data. As a result, it is critical to make sure that an assessment framework encourages evidence-based practice while also taking into account the various types of evidence. An evidence-based assessment framework, for example, should contain particular data collecting items and compare the acquired data to a body of literature to corroborate assessment conclusions. The importance of standardizing is highlighted by frameworks having such qualities.

Research question
The following research question informed the study: What preferred common factors and good practices underpin current foster parent screening and assessment frameworks used by agencies in the North-West Province?

Approach
This study adopted a quantitative approach primarily guided by an evaluation paradigm in a descriptive and exploratory manner. This approach enabled the researchers to establish the content components of different assessment frameworks used by several welfare organisations in the North West Province, South Africa, and provide recommendations for future development and standardisation purposes. Furthermore, because the researchers intended users to rate and evaluate the frameworks, the evaluation criteria served as a starting point for further analysis to determine the degree of (1) relevancy and (2) appropriateness of the assessment frameworks offered. Both criteria focused on substance, such as assessment dimensions and whether or not they were relevant and sufficient. There were two stages to the research. The collecting of several assessment frameworks from selected welfare organizations was phase one of this investigation. The second phase included a workshop with social workers from several welfare organizations, followed by an evaluation of the various assessment frameworks. For this purpose, respondents were requested to indicate using rating the content validity of existing assessment frameworks.
We divided the scale elements into three quality dimensions for this study: (1) functionality, (2) technicality, and (3) usability. The requirements for each of the five assessment frameworks were represented in each section. The rating instrument was created for data collection throughout this phase. The method was interested in how the assessment frameworks appeared to the respondents enabling us to determine the face validity of the five assessment frameworks (Krabbe, 2017).

Design
The above-mentioned strategy was supported by the employment of a cross-sectional survey design (Creswell, 2014). We used this design for quantitative data collection and aimed to evaluate the different frameworks by means of interrater agreement to determine the extent to which each assessment frameworks' content represented the raters' expectations using a fully-crossed design. The researchers were able to assess the level of agreement between two or more responders with the use of the fully-crossed design. The inter-rater agreement of the five specified assessment frameworks was evaluated using a self-developed, paper-based survey using a 5-point Likert scale.

Respondents
The respondents in this study represented several non-governmental organizations (NGOs) in the North West Province of South Africa, and included two districts, Dr. Kenneth Kaunda and Dr. Ruth Segomotsi Mompati. We disseminated recruiting leaflets and communication to all child-protection organizations in the respective districts, and four of them, CWSA, NG Welfare, RATA Social Solutions, and SAVF, chose to participate. The demographic statistics in the Table  below shows the respondents per district:

Sampling
This study consisted of two phases, with the first phase employing a two-stage sample technique that included deliberate sampling. The goal of the first phase was to identify and collect existing evaluation frameworks used by various welfare organizations for screening and assessing potential foster parents (Rubin & Babbie, 2013). The researchers sought and chose designated welfare organizations in the North-West Province from the two mentioned districts to form the sample.
Phase 2 of the study consisted of a data collection day at the North-West University during which representatives of the different child protection organisations in the province presented their assessment frameworks to the respondents. In addition, a lecture presentation on scientifically supported assessment strategies and how to evaluate frameworks for adequacy and relevance provided the necessary education to respondents in preparation for the project as a direct benefit of the research. The availability sampling approach was utilized for phase two (2) of the study, which comprised individual social workers from the welfare organizations that participated in phase one (1). The researchers wanted at least two social workers and one director from each organization to take part in the study. They thought this group would be sufficient for explaining how to utilize and apply each assessment framework.
A sample of at least 40 people would was aimed for. However, the researchers were unable to achieve the desired sample size since most of the organizations involved were small offices, and social workers could not afford to take a day off and travel long distances to participate in the study. Despite these constraints, 18 participants attended the session. The study's approach for obtaining rater-agreement levels were not sensitive to sample size and hence had no impact on the study's conclusion (Donner & Rotondi, 2010).

Data collection instrument
Whereas Phase 1 of the study employed a relatively simple invitation and questionnaire with a request to submit the organisation's assessment framework for scrutiny and evaluation, phase 2 relied on a rating scale that the researchers developed for purposes of the study. Literature guided the rating scale item development and resulted in a scale measuring three dimensions: Assessment objectives (Functionality), Technical properties (Technicality) and Utility value. Each dimension had 7 -9 items, and a five-point Likert scale where 1 = Strongly disagree and 5 = Strongly agree was used. Each dimension consisted of five items. Recruited respondents attended the data collection workshop, where they performed the ratings after the orientation session. The workshop also presented the benefits of using assessment frameworks in practice and differentiating between the properties of assessment frameworks versus flawed assessment frameworks based on their content. Finally, training focused on using the rating scale and conducting an analysis of each sample assessment framework. The workshop assisted respondents in making well-informed decisions regarding the evaluation and rating of the assessment frameworks to improve the study's validity. They then studied the five samples of assessment frameworks issued to them and rate these using the refined version of the rating scale. Thus, each respondent evaluated and rated all five (5) assessment frameworks (Hallgren, 2012). Lastly, the survey also allowed the respondents to rate the assessment frameworks according to their relevancy and adequacy.

Validity and reliability of the rating scales
Prior to utilising any scale, users need to take cognisance of its psychometric properties (Tay & Jebb, 2017). Since the researchers designed the rating scale used in this study themselves, it lacked any psychometric properties and the researchers at the very least needed to confirm its face and content validity. In this instance, the researchers qualitatively established the content validity of the data collection instrument by utilising consultation meetings amongst the researchers and volunteer colleagues at the school of social work in which consensus about item formulations formed the basis of scale validity. Secondly, the North-West University Statistical Services and an expert in quantitative research researchers determined the face validity of the survey. Finally, the scale's validity was determined by the consistency of the material contained in the assessment framework, as well as their potential to be altered and used in a variety of practice settings (Kimberlin & Winterstein, 2008). Due to the small sample size, it was not possible to evaluate scale validity, but reliability was established empirically using Cronbach's Alpha.

Method of data collection
The researchers used two different data sources. The first data source was the five evaluation frameworks the organisations used to screen and assess potential foster parents. These frameworks were samples of assessment frameworks and were selected during phase one (1) of the study and rendered a selection of frameworks and procedures compiled as a portfolio of evidence representing the most prominent frameworks used. Next, the researchers evaluated these portfolios against pre-designed quality specifications that enabled them to assess whether these frameworks met the minimum criteria for inclusion in phase two (2) of the study. Rating scales were completed on paper and the data was then captured on an IBM SPSS database in preparation for analysis.

Method of data analysis
The objective of the analysis was to determine what aspects of each framework could be considered strengths or weaknesses. The analysis outcome would be recommendations regarding those features in a framework that contribute to high clarity, technical strength and utility value. Data retrieved from the survey were analysed using the IBM SPSS Statistics Version 25, Release 25.0 (SPSS) and a North-West University Statistical Services statistician. The analysis delivered descriptive and frequency results that described the basic profile of the sample. The researchers then evaluated the reliability of each section of the survey by utilising Cronbach's alpha. The descriptive statistics' mean scores allowed the researchers to perform useful comparisons between specific statements and assessment frameworks, as well as identify the most and least beneficial frameworks in terms of the dimensions (B) aim clarity, (C) technical qualities, and (D) utility value. The analysis consisted primarily of evaluating the Mean score per item to indicate the amount of support for a framework dimension. Since the five-point scale reflected progressively positive expressions regarding the framework, it was possible to decide on the highest-scoring frameworks.

Respondents identifying information
The respondents provided the following demographic data for the study: (1) gender, (2) age, (3) educational qualification, (4) years of experience as a social worker, (5) years of screening and assessing potential foster parents, and (6) the name of the organization where they worked.
The majority of respondents were female, and more than two-thirds of the sample were younger than 45 years of age, as indicated in the table below. Ten respondents had up to 12 years of experience as social workers, and the remaining 7 were senior workers with 13 -26 years of experience. The experience level was in foster care screening which was appropriate for this kind of study. The majority of those who responded had little to no experience screening and assessing potential foster parents.

The openness of respondents to adapt their assessment frameworks
The willingness of respondents to adapt their organization's evaluation framework reflected the degree of change required in their organizations. There was a high level of willingness to modify their organization's evaluation framework, with 88.9% of respondents eager to do so, and only 5.6 percent unwilling to do so. The question was left unanswered by one respondent. This study implies that the majority of practitioners want to examine and revise their frameworks in order to better their profession.

Statistical reliability
All three dimensions of the self-compiled rating instrument and each of the five assessment frameworks (A -E) have its own set of reliability data. The survey questionnaire assessed the five assessment frameworks' distinct content needs and provided the researchers with an estimate of instrument performance on each framework. Below are Cronbach alpha values and descriptive data from the various assessment frameworks, as well as the three parts. Cronbach's alpha should be between 0.5 and 0.9 as a recommended value for assessment systems (Mohajan, 2017). Based on the Cronbach Alpha, each framework's performance was evaluated. The researchers considered low Cronbach alpha values below the guideline range, as suggesting inconsistency or doubt in assessments regarding that frameworks. Second, the researcher looked at how respondents rated the various assessment systems.  As indicated in the above table, reported Cronbach alphas indicated that all raters rated consistently and within the range of guide values proposed above. Thus, we concluded that the survey questionnaire achieved the reliability standard across frameworks and dimensions. However, we were concerned that specific sections within a framework were rated less reliably than other sections. For example, Section B of Framework E returned a low alpha ( = 0.688), indicating that this section was not very clear in that framework. Similarly, Sections C in Framework A, C and D were also less reliably answered ( = 0.653, 0.608 and 0.576). Therefore, we interpreted these sections with greater caution.

Frequency and descriptive statistics
For each of the five assessment frameworks (Assessment framework A -E), frequency and descriptive statistics were presented individually. It allowed the researcher to draw conclusions about variable distribution and find key trends in the data.

Comparison between assessment frameworks
The next three parts were accompanied with results, which included the mean and standard deviation of the various evaluation frameworks: (B) Framework objectives; (C) Technical standards and (D) Utility value, are represented below. The results provided item Means and Standard Deviations and the overall Mean, Standard Deviation (SD) framework ranking based on the overall Mean across sections. We discuss these according to each section. Based on the above results, we compiled a ranking table representing choices in frameworks as indicated in the following Table   Table 6: Primary ranking table per framework

Framework Ranking
The ranking of the five evaluation frameworks was decided using mean scores and standard deviations. Assessment Framework E, which was closely followed by Assessment Framework A, accomplished the most framework objectives. The greatest outcomes were seen in Assessment Framework E, particularly in terms of having a defined assessment scope and assessing utilizing several data sources. Assessment Framework A, on the other hand, received the highest ratings for having a clear purpose and being a formal document. Multiple scale items in Assessment Frameworks B and D had a significant standard deviation (above 1), suggesting greater variation in respondent ratings. However, the highest standard deviation (SD=1.237) for using a specific technique for evaluation or scoring (scale item 8) was found in assessment Framework D, indicating that respondents were unsure if this framework matched the underlying method and scoring criteria. When interpreting the total SD, we discovered that framework D had the largest variability, implying that its attributes are less definite and clear to raters.  The overall ranking based on Mean scores for utility value was as follows: Table 9: Overall rankings for Utility value We tabled all framework rankings across the three dimensions based on the above results to evaluate each frameworks' strengths and weaknesses. The ranking table gave the researchers an index of framework acceptability relative to the different dimensions of ratings. Thus, it became possible to assess what weaknesses to avoid in future framework selection. Assessment Framework A was ranked first, followed by Assessment Framework E, Assessment Framework C, Assessment Framework D, and Assessment Framework B. We then attempted to confirm the final recommended framework by evaluating the Standard deviations as a measure of variation in ratings. Lesser standard deviations, we reasoned, indicated a better degree of rating consistency, whereas higher standard deviations suggested a lower level of uniformity and greater variation. Although some of the less desirable frameworks returned higher SD's and, by implication, reflected more varied responses, we interpreted this as more uncertain, varied ratings, although the pattern was not consistent for all frameworks.

DISCUSSION
The findings show a lack of uniformity in using assessment frameworks during the screening and assessment of prospective foster parents, as variations in sample frameworks according to the quantitative section of this study showed. This variation was because sample frameworks were mainly developed in-house to different welfare organizations and based on their practical requirements as they saw fit. As a result, frameworks were non-standardised and customised to a level of convenience rather than scientific rigour. Furthermore, although organisations used collective wisdom in developing these frameworks, frameworks were not necessarily subjected to scientific scrutiny and best practice evaluation. Respondents also expressed the desire to change their organisation's screening and assessment frameworks and consider the usage of standardised assessment frameworks that captured the essential features more uniformly.
Furthermore, the findings of this study showed that the assessment frameworks included in this study lacked systematic standardisation processes. Therefore, this study suggests that the ideal assessment framework has to meet specific requirements concerning its functionality, technicality, and utility as these appeared to be central to improving assessment frameworks.

LIMITATIONS
The study included the following limitations: The data collecting technique used in this study was a self-developed paper-based survey questionnaire to quantify the degree of agreement or disagreement between respondents. This methodology was feasible for this study, but it was impossible to validate the study's instrument. Unfortunately, the researchers were only able to recruit 18 respondents to participate in the study, which limited the analysis possibilities as a larger data set of 40 would have been preferable and would have thoroughly explained high levels of variability as demonstrated by big SD values.
The demarcation of the study to two districts in one province limited the study findings somewhat. A more extensive study including other districts might have resulted in better participation and more substantive evaluation. In addition, respondents to the research study included non-governmental organizations that provide foster care services. As a result, the conclusions of this study cannot be applied to the whole population or organizations providing social services.

RECOMMENDATIONS FOR PRACTICE
Based on the recommendations of respondents after completing the ratings they agreed to promote the implementation and use of uniform standardised assessment frameworks in the screening and assessment of prospective foster parents. Specifically, in South Africa, such a universal framework is presently far from being developed, as more research is needed to explore further instrument usage and assessment procedures in other contexts and districts. Therefore, it is impossible to recommend a single instrument but rather understand the different elements that make up a constructive framework with high utility value. However, this study has shown at a small scale what features users prefer and what essential features require inclusion in a framework that will improve its implementation and adoption. It would be advisable to evaluate frameworks and other assessment methods such as interviews styles, home visits, and report writing formats to improve the number of data sources for decision-making and the overall utility value of the framework. There is a need to standardise practices that eventually contribute to better quality practices and services to clients.

RECOMMENDATIONS FOR FUTURE STUDIES
Given that the instrument has shown to be trustworthy and appropriate for its intended use, it now requires further improvement in order to examine new and alternative evaluation frameworks.
Replicating this study with a bigger population would start a more thorough review of frameworks, eventually leading to the selection of a framework with better attributes and features.

COMPLIANCE WITH ETHICAL STANDARDS
As set out by the North-West University Health Research Ethics Committee, this study followed ethical compliance and was approved with an ethical reference number: NWU-00016-18-S1.