Experiences of experts in intelligence measurement of South African school learners

of intelligence of South African school learners in diverse contexts. Findings of this research study can guide the strategic process to design an intelligence instrument suitable for a South African population of school learners, informing fair assessment practices for multiethnic equalisation.


Introduction
Influenced by the post-apartheid legislation promulgated in the Bill of Rights of South Africa as part of the Constitution (South African Government, 1996), South African practitioners in the field of the psychometric assessment of intelligence of school learners and psychometric instrument adaptation and/or development realised the need and obligation for fair assessment practices (Foxcroft et al., 2004;Laher & Cockcroft, 2014;Meiring, 2007).This would include the administration, creation, or adaptation of fair and unbiased assessment instruments applicable to a multilingual and multiethnic society (Laher & Cockcroft, 2014).The South African school learner population is diverse, not only in terms of culture, ethnicity, and language, but also because of past unequal socio-economic and educational opportunities (Laher & Cockcroft, 2017;Shuttleworth-Edwards, 2016;Van der Merwe et al., 2022).Despite almost three decades of restoration practices in the democratic South Africa, vast discrepancies remain among the school learners (National Planning Commission [NPC], 2020; Ogbonnaya & Awuah, 2019;Van der Berg et al., 2020).There are different population sub-groups based on socio-economic variances, resulting in school learners having unequal access to essential services and disparate levels of quality environments found both at home and in school (Laher & Cockcroft, 2017;NPC, 2020;Ogbonnaya & Awuah, 2019;Van der Berg et al., 2020).Assessing the intelligence of multilingual and multiethnic local school learners from and within various environments poses challenges to both experts developing and/or adapting intelligence instruments, and psychometrists and/or psychologists administering the instruments (Foxcroft, 2004;Laher & Cockcroft, 2017;Lucas, 2013;Van der Merwe et al., 2022).According to the critical review conducted by Van der Merwe et al. (2022), there is no specific reference made to published studies in South Africa on various tests that assess intellectual functioning of South African school learners.The current manuscript then asks the further question regarding what those using these tests say about their experiences.This qualitative research study emerged from the essential need for reliable and valid intelligence test instruments for South African school learners, who are characterised as a diverse population with their variety in culture, ethnicity, and language, as well as having unequal socio-economic and educational backgrounds.The aim of this research study was to use a qualitative interpretive description research design to explore and describe the experiences of both experts in intelligence test development and/or adaptation as well as psychologists and psychometrists who have administered intelligence tests to South African school learners in various contexts.Twelve psychologists and/or psychometrists were interviewed, of which six were also experts in test development and/or adaptation, which yielded four themes after thematic analysis, namely, utilised intelligence measurements in the current South African school learner context are less relevant; the South African education system is a major issue specifically within lower socio-economic status (SES) contexts; it does not seem feasible to design or adapt suitable intelligence measures that are valid and reliable in the current South African school learner context; and key informants' recommendations from their experiences.

Historical changes in the new democratic South Africa
In 2004, a decade after the commencement of the democratic South Africa, the Human Sciences Research Council (HSRC) conducted a comprehensive survey into the test-use patterns and needs of psychological assessment practitioners in South Africa (Foxcroft et al., 2004).Key findings included that many psychometric tests were outdated and not linguistically or cross-culturally relevant, consequently reducing their value, leading to an urgent plea for reliable revision and adjustments (Foxcroft et al., 2004).From the beginning of the democratic South Africa in 1994, a number of normative research projects were launched on differing population groups using various assessment instruments (August, 2017;Laher & Cockcroft, 2017;Laher et al., 2019).According to Van der Merwe et al. (2022), contextual and demographic factors, for example quality of schooling, language, and culture, could affect the school learners' test performance and consequent intervention plans, which could prove ineffective if the test instrument was outdated and not valid and reliable for the group of learners assessed.

Present South African intelligence assessment environment
South African psychologists and psychometrists presently have a limited selection of individual intelligence tests for assessing school learners (Mitchell et al., 2018;Van der Merwe et al., 2022).These instruments include the Learning Potential Computerised Adaptive Test (LPCAT;De Beer, 2000), Senior South African Individual Scale-Revised (SSAIS-R; Van Eeden, 1991), and Junior South African Individual Scale (JSAIS; Madge, 1981), which are locally developed tests (Van der Merwe et al., 2022).As a result of the scarcity of such locally developed instruments, South African practitioners frequently need to rely on imported tests that are expensive and not developed to meet the needs of local populations (Laher & Cockcroft, 2017;Shuttleworth-Edwards, 2016).The Raven's Educational Progressive Matrices and Vocabulary Scales (Raven, 2008), Kaufman Assessment Batteries for Children, 2nd edition (KABC-II; Kaufman & Kaufman, 2004), and the Wechsler Intelligence Scale for Children, 5th edition (WISC-V; Wechsler, 2014) serve as examples of such tests administered in South Africa (Benadé, 2023;Cassoojee, 2020).
The intelligence test instruments commonly administered in South Africa are based on the psychometric model of testing, determining intelligence according to the individual's ability to perform test tasks and providing an indication of cognitive strengths and weaknesses (Cockcroft, 2013).To conduct such tests effectively as psychologists and/or psychometrists in a new inclusive framework, as promulgated by South African legislative policies, Swart and Pettipher (2005) propose shifting the approach from a deficient-focused assessment (as prescribed by the medical deficit model) towards an assetbased, social systems change approach, where the school learner's personal strengths (intrinsic qualities) and assets (extrinsic resources in their social environment) are identified and utilised.The psychological assessment practice should be conducted in a dynamic and flexible manner, selecting 'a battery of assessment measures and procedures applicable to the needs of the specific client and in response to the specific referral question' (Elkonin et al., 2004, p. 282).This, however, is not entirely possible if most assessment instruments available to these experts are not applicable or outdated (Lucas, 2013;Mitchell et al., 2018;Van der Merwe et al., 2022).Key findings from the critical review by Van der Merwe et al. (2022) reveal the inapplicability of many intelligence instruments when administered to diverse South African school learners.This renders them invalid, inappropriate, and unfair for many school learners.Contextual and demographic influences can impact significantly on test performances, especially if the test was originally designed for and normed using a different population group or developed within a different time frame.This underscores the importance of considering the test instrument's content, developed norms, and the date of publication before assessing learners (Van der Merwe et al., 2022).Moreover, there is a lack of published studies that cover the topic of applying various tests that appropriately assess intellectual functioning of school learners in South Africa (Van der Merwe et al., 2022).
In light of the above-mentioned statements, this research study aimed to address this gap in the field of psychometry, by collecting data from the experiences of a group of professionals working within the multicontextual domain of the measurement of intelligence of South African school learners.The following research question guided this study: What are the experiences of intelligence test developing and adaptation experts as well as psychologists and psychometrists who have administered intelligence assessment instruments to South African school learners in various contexts?

Sample
Two types of population groups were included in the research study.The first consisted of professional experts with knowledge and experience in developing and/or adapting intelligence tests for reasons such as standardised norming, updating, or adapting.They should have adapted and/or developed at least one intelligence test, as they needed to have experienced this process to provide any information regarding the research topic.Psychologists and psychometrists registered at the Health Professions Council of South Africa (HPCSA) and practising in the field of psychometric testing were chosen as the second group type.They needed to have sufficient experience in conducting intellectual assessments on school learners in various contexts in South Africa.The final sample consisted of 12 psychologists and/or psychometrists, of which six were also experts in test development and/or adaptation.Participation in the research study was voluntary.Participants were diverse in terms of gender, race, type of HPCSA registration and/or field of practice, and number of years practising in their registered fields (see Table 1).As illustrated in Table 2, the participants represented all nine provinces across South Africa.Their experience in the field of psychology and/or psychometry varied.Collectively, the participants have experienced working with school learners from the five mainly found race groups in South Africa, who attend schooling in the three phases found in South African schools, namely Foundation, Intermediate, and Senior Phase.Most of the participants worked with learners from all four levels of SES.Learners' parents varied in their levels of education received, from having no scholastic background up to having a tertiary level of education (see Table 2).

Data collection
Data were collected through a demographic questionnaire and semi-structured interviews, using an internet-based communication platform.Semi-structured interviews enabled the primary researcher to present herself in a formal, wellprepared manner, as well as leave room for probing and clarifying questions during the interviews.This led to flexible, interactive, and responsive conversations, allowing for qualitative, meaningful, and in-depth data (Williams, 2019).Questions were asked in the participants' preferred language (English or Afrikaans; see Table 3

[English version only]).
Interviews were held until data saturation was reached.

Data analysis
The interviews held were transcribed verbatim, from which themes were identified through reflexive thematic analysis (Braun & Clarke, 2021;Byrne, 2022), following procedures proposed by Clarke and Braun (2013): (1) familiarisation process by actively reading and re-reading through the data with interpretive reflection; (2) assigning code labels to text segments according to meaning and significance; (3) identifying themes from codes; (4) reviewing themes against the research question; (5) defining and labelling themes; and (6) writing a final report.Coding was independently conducted by the first and third authors, as well as a registered research psychology intern within the School of Psychosocial Health, North-West University, acting as independent, external data coder and analyser to interpret data with a neutral, unbiased perception.This process was followed by a meeting held by the coders and second author, where notes (visual illustrations of each coder's identified codes and suggested themes) were compared and discussed to determine the final themes and findings.
Taking the data analysis process of constructing and handling data in relation to the research topic to a second, more progressive level, as suggested by the interpretive description design, the researchers aspired to go beyond merely jotting down a list of identified topics.The data analysts employed sound inductive reasoning skills by studying, reflecting, and questioning identified elements interpretively and critically within the context of their relationship to each other, as well as weighing up and confirming findings against the research topic, context, and theoretical knowledge base, as suggested by Thorne (2016).

Results
Four main themes with sub-themes emerged from the analysed recorded data (see Table 4).To support the themes, applicable verbatim quotations are provided.

Theme 1: Utilised intelligence measurements in the current South African school learner context are less relevant
From the collected data, it emerged that the intelligence measurements being administered to South African school learners have become outdated and inapplicable when considering their present, modern, multicultural, and multilingual contexts.The absence of continuous research seemed to be an important explanation.

Sub-theme 1.1: A lack of ongoing research
Participants commented on the utilised intelligence instruments being outdated and not applicable in the school learners' present contexts because ongoing research did not occur.Participant 5 stated: Hence, it appears that intelligence measurements administered to South African school learners have become less relevant when considering the learners' current contexts, because continuous research did not happen.

Sub-theme 1.2: Due to multilingual and multicultural contexts
Many participants ascribed the unsuitability and inappropriateness of many utilised intelligence instruments to the multilingual and multicultural formation of local school learner groups.Participant 6, responding to whether locally administered tests are applicable to local school learners: 'Generally, no, and that is the dilemma that we have   She also stated how language could act as a barrier if the test is not administered in the school learner's first language:

Partici pant Numbers
'[…] the language of test administration, if it's not the first language of the learner, obviously, that's a barrier.' (P6, Female, Research Psychologist) Some participants stated how tests cannot be translated into the 11 official South African languages because of various existing dialects and obstacles with translations and backtranslations.Other participants described how certain concepts of the same language differ, depending on the area in which that language is spoken.Some participants found that the role of ethnicity differences between assessor and assessee could also influence assessments.
It would appear that the utilised intelligence instruments are not appropriate for many local school learners, specifically when considering learners' contextual backgrounds that are linguistically and culturally diverse.

Sub-theme 1.3: Outdated items
Participants mentioned that items in intelligence instruments, including language and graphics, were outdated and thus irrelevant to the school learner's modern world.Participants 8 and 12 explained that certain items in the SSAIS-R and JSAIS are outdated, making them unsuitable to the modern culture:  Thus, items found in intelligence tests are outdated, making them unsuitable, less relevant, and thus questionable to administer to local school learners in their modern contexts.Apart from this, as is discussed in the next theme, it seems as though the country's education system acts as a major contributor to the issues found in intelligence testing.
Theme 2: The South African education system is a major issue specifically within lower socioeconomic status contexts Lower quality of education provided by schools located in the lower socio-economic status (SES) areas was indicated as contributing to the issue of intelligence instrument inapplicability.

Sub-theme 2.1: Many schools within lower socioeconomic status contexts in South Africa provide substandard education for school learners
Many schools located in lower socio-economic environments were reported to provide lower quality education.This may act as a hinderance that causes these school learners to perform inadequately on intelligence tests.Participant 12 mentioned: In view of the hindering influence of varied educational provision on school learner intellectual test performance, participants commented on how results from these measurements need to be carefully considered.

Sub-theme 2.2: Should therefore be careful how we interpret results of current intelligence measures
Considering fair assessment practice and learners' educational backgrounds, Participant 6 mentioned how school learners' scores on intelligence tests need to be interpreted cautiously: 'We must be very careful how we interpret results, unless we can really motivate that it is a fair way of assessing […]  Participant 5 cautioned that school learners from rural backgrounds may perform below their potential when answering test questions because of limited exposure to the testing language of English: '"[…] with your rural and shy learner that does not necessarily mean that they don't know when they say they don't know, so you probe a bit more […]" because, "[…] a child who's not exposed to speaking in English all the way [speaking English sufficiently], will answer yes or no when it's not necessarily the correct answer".' (P5, Female, Educational Psychologist) Because of differing information exposure provided in various environments as well as the verbal element found in intelligence tests, Participant 7 claimed that one cannot infer lower cognitive functioning from underperformance in tests: '[…] there're some of the words in the Vocabulary subtest which children are not exposed to.So, if they cannot answer it, it is not necessarily an indication of lower cognitive functioning, it's [rather] an indication of not being part of their world of living.' (P7, Female, Educational Psychologist) Therefore, substandard education provided by many schools within lower SES contexts may act as a major contributor to the school learners' inadequate performance on intelligence tests, which necessitates the careful interpretation of these school learners' performance results.Considering the present school environments, it may seem unfeasible to standardise new or adapted intelligence test measurements.

Theme 3: It does not seem feasible to design or adapt suitable intelligence measures that are valid and reliable in the current South African school learner context
Based upon findings stipulated in themes 1 and 2, attempting to design and adapt valid and reliable tests that will accommodate each and every school learner in the various South African contexts, seemed unfeasible.This perception was supported by Participant 6: Some participants observed the loss of the governmentsupported HSRC not performing the role of psychometric test development anymore and questioned the possibility of really being able to collect representative samples.
The above-mentioned challenges suggest that intelligence measurements cannot really be developed or adapted to suit the school learners from diverse areas and educational backgrounds in South Africa.In the next theme, key participants have provided some helpful suggestions of how one could go about still attempting this seemingly impossible task.

Theme 4: Key informants' recommendations from their experiences
Derived from their experiences of developing, adapting and/ or administering intelligence test batteries to school learners in various regions of South Africa, key participants have shared significant ideas of how to approach the challenging task of attempting to develop and/or adapt intelligence measurements that are fair and applicable.

Sub-theme 4.1: Need to develop our own South African framework
Key participants mentioned the need to create a theoretical framework for intelligence testing that is authentic to South Africa, upon which fair and applicable intelligence measurements can be built.Calling it 'a unique construct', Participant 6 encouraged designing an original, personalised framework: '"We need to go back to the clean slate drawing board and find out how is intelligence experienced, seen, defined, observed, in a multicultural South African context […]" and "[…] from there, try to operationalise this into something measurable and valuable […] in a way that everybody can agree is representative of what is considered intelligent behaviour, or intelligent reasoning".' (P6, Female, Research Psychologist) With tolerance and acceptance, over time, intelligence could be uniquely defined and conceptualised through the multidimensional viewpoints of various South Africans joining forces.

Sub-theme 4.3: Need to base use of type of intelligence measurement on context or functioning of the school learner
Key informants recommended selecting administered intelligence tests according to the school learner's environment or behaviour.Participant 3 expressed this well by stating that: Participant 3 explained that the assessment aim determines the test being selected: '"[…] there's a range of intelligence assessments […] [which] measure different things […] sometimes just a person's GMA and other times a far more comprehensive, complex understanding of their reasoning [ability] […]" therefor, "[…] all of that comes into play in how we select our tools".'(P3, Female, Psychometrist and Research Psychologist) She noticed, however, how practitioners are limited by only being allowed to administer tests recognised by the HPCSA: '"Selecting the tools that I use need to be on the HPCSA list […]" however, "[…] sometimes we have assessments […] with [adequate] sample size, reliability, validity, non-bias […] differential item functioning and all of those fun things, but it's not on the HPCSA list, so I can't use it".' (P3, Female, Psychometrist and Research Psychologist) Reflecting on the above-mentioned statements, it became clear that school learners' contexts and functioning need to be considered when selecting intelligence instruments for fair assessment.As these learners reside in modern contexts, intelligence instruments should also replicate this essential feature.

Sub-theme 4.4: Update to reflect modern context
To represent the modern elements of South African school learners' contexts, intelligence tests need to be updated.Participant 1 explained that modernising intelligence instruments: '"scientifically and statistically" should make them applicable again: "[…] adaptations and renewal […] [to] be applicable again […] a new instrument that is scientifically and statistically up to standard".'(P1, Male, Psychometrist) The following participants described how updated versions need to reflect the school learners' modern contexts: Modernising intelligence instruments includes updating norms, which need to be performed in the correct manner.Participant 9 explained how subgroup norming could be applied to heterogenic groups; however, the context needs to have fewer variables:

Sub-theme 4.5: Appropriate norming needed
'If the population in terms of demographic factors, education, socio-economics, language, etc. is heterogenic […] use subgroup norms, but it only works within very specific environments, the use of within-group norms.' (P9,Female,Psychometrist and Research Psychologist) Other than correct norming practices, key participants also recommended the creation of optimal contexts and practices during intelligence assessment.
Sub-theme 4.6: Need to establish optimal contexts and practices during preparation, assessment, and reporting processes Practitioners need to ensure favourable contexts and practices before, during, and after the assessment process.Participant 6 mentioned how practitioners should ensure a fair and appropriate assessment process: To address the stigma that is formed from these unfair practices, Participants 6, 8, and 11 recommended taking time to explain to the school learners of the features and practices involved in the assessment process.
On the day of the assessment, some participants mentioned checking if the children have travelled far before assessment or had enough sleep and nutrition, and attempted to support where possible to provide them an opportunity for best performance.Other than basic needs, Participant 7 noticed the importance of drawing attention to challenges found in school learners' contexts, which could affect their performance: '"I would ask a child: 'How did you get to school?'" because, for instance, "If a child has driven with an over-anxious, activated, aggressive parent before the assessment, you will sit with a child experiencing anxiousness […] those previous experiences should be taken into account".'(P7, Female, Educational Psychologist) Lastly, some participants recommended that the measurement of intelligence not solely rely on scores derived from the intelligence test, but should include a battery of measures and other information gathered from interviews, observation, and more:  (Foxcroft et al., 2004;Tredoux, 2013).Presently, intelligence test instruments have become outdated, inapplicable, and unfair to administer to multicultural and multilingual school learners within various contexts (Laher et al., 2019;Shuttleworth-Edwards, 2023;Van der Merwe et al., 2022); however, these tests are 'more reliable and valid than any of the limited number of alternatives' (Laher & Cockcroft, 2013, p. 4).Mitchell et al. (2018) mention the shortage of available culturally appropriate cognitive tests for school learners.This places practitioners in the field of psychometric testing in a challenging predicament, as they are legally and ethically bound to conduct fair and valid testing for all (HPCSA, 2016;ITC, 2015;South African Government, 1996).
Many participants indicated that the unsuitability of intelligence tests is owed to the substandard education provided, specifically by schools located within lower SES contexts, which affect the ability and performance of school learners from these schools.Careful interpretation of the school learners' test scores and performances was accordingly argued, acknowledging differing exposures to language and information.This agrees with Shuttleworth-Edwards et al. ( 2013), who stated how the quality of education received by local school learners is related to their performance on intelligence tests.Various South African normative studies posited that considerable differences in intelligence test scores of school learners were not primarily because of ethnic differences, but rather the differences in the quality of education (Amod, 2013;Greenop et al., 2013;Shuttleworth-Edwards et al., 2013).Clearly, South African schools provide various levels in quality of education when comparing the five quintiles into which schools are categorised, according to their SES (Maistry & Africa, 2020;Ogbonnaya & Awuah, 2019) (see quintile and poverty distribution across provinces in DBE [2021]).School learners who go to schools located in disadvantaged areas (quintiles one to three) tend to receive lower quality of education and yield lower academic performances than school learners who attend more privileged schools (quintiles four and five) (Graven, 2014;Ogbonnaya & Awuah, 2019;Spaull, 2012).Inequalities in education provided across various schools have been noticed (see Spaull & Jansen 2019).Referring to the SACMEQ III dataset for South Africa, Spaull (2012) remarked that a school's SES had a far greater impact on school learner performance than individual learner's socio-economic background.In addition, Mitchell et al. (2018, p. 435) have drawn attention to how the locally developed intelligence test battery, namely the SSAIS-R, is 'somewhat biased to educational exposure, in particular verbal and linguistic abilities', which further supports the argument for unfair, inapplicable intelligence measuring instruments where careful interpretation of intelligence measured results is needed, especially of school learners who attend schools within lower socio-economic contexts (Laher et al., 2019;Mitchell et al., 2018;Shuttleworth-Edwards, 2023).
Guided by the Constitution (South African Government, 1996), as well as guidelines provided by the HPCSA (HPCSA, 2016) and the International Test Commission (ITC, 2015), practitioners in the field of psychometric testing are legally, ethically, and professionally obligated to not only administer measurements that are valid, reliable, and fair to all but also to not withhold services because of reasons of ethnicity, language, culture, among others.As participants regarded the heterogenic nature of the local school learners with their differing contexts, it seemed unrealistic that newly or adapted intelligence measures could be applicable to all South African learners.Examples of challenges include translation difficulties (particularly with multiple South African languages with different vocabularies and language rules to English, and additionally the existence of various regional dialects), issues with test norming and standardisation practices within multiethnic population groups, as well as limited research opportunities because of a lack of resources and accessibility.Questions and concerns were raised around topics such as the influence of demographics and the Flynneffects during norm development, the possibility of conducting fair and appropriate comparative norming practices within multidimensional population groups, and the feasibility of developing novel, African-centred intelligence tests.Standardisation challenges and concerns were also raised by authors such as Aston (2006) andTruter et al. (2018), who have mentioned how the population-based projects of standardising the Wechsler's scales (third and fourth editions) for the whole South African population, have received criticism regarding the norming processes followed.Shuttleworth-Edwards (2016, 2017) states that practices of country-wide unitarity norming applied on intelligence tests in countries with culturally heterogenous contexts are invalid, as South Africa does not represent one unified population culture that could be generalised.Shuttleworth-Edwards and Truter (2023) present compelling arguments concerning the practical utilisation of contextspecific, within-group norms (in contrast to populationgroup norms) within a diverse, heterogeneous population across various contexts.Laher and Cockcroft (2014) highlight the positive strides made in developing (e.g., LPCAT), adapting, norming, and standardising psychological test instruments suitable for local contexts; however, they highlighted the need for developing emic measurements that are appropriate for the local African population for whom Western-developed tests may be inapplicable.
Intelligence tests and instruments still function as essential means towards school learner guidance and support, where the combination of interpreted measured cognitive abilities (e.g., their strengths and challenges) and other findings (e.g., learner background information or observations from testing) aid the practitioner in developing applicable intervention strategies (Beal et al., 2019).Despite the former concerns and challenges regarding fair intelligence testing of local school learners, participants have shared and provided insightful suggestions of how one could undertake this seemingly unfeasible task.Recommendations included gathering input from various local groups to define and conceptualise the term 'intelligence', designing an original, personalised South African framework, aligning the intelligence measurement to the school learners' context and functioning, updating the intelligence instrument to reflect the learners' contexts, norming the intelligence tests adequately, as well as ensuring optimal contexts and and dependability (keeping an audit trail, using co-coders, and taking guidance from reviewers).

'
[…]  we see that they often function on a low level[…]   scholastically they show much more handicaps than our kids who might come from another socio-economic group.' (P12, Female, Educational Psychologist) Participant 8 noticed the poor quality of schooling offered by schools in disadvantaged areas: '"[…] about 90% of them come from a more disadvantaged socio-economic background […] often with very poor quality of schooling […] either in a township or rural area" and, "[…] what I mean by disadvantaged is that the school itself is not well resourced".' (P8, Female, Counselling and Neuropsychologist) Providing examples: '[…] they don't have libraries or computer rooms, often toilets aren't working properly.Teachers are often not trained all that well and not all that motivated to teach, classes are overfull …' (P8, Female, Counselling and Neuropsychologist)Other challenges:'"[…] often children walk very far to school and it's dangerous, so they're not always motivated to attend school every day […]" also, "[…] parents are often not well educated themselves, so can't help them with their schooling, don't really see the need for schooling […]" and, "[…] not really all that involved in their [children's] lives […] a lot of drug abuse, alcohol abuse […]".' (P8, Female, Counselling and Neuropsychologist)

'
[…]  a unique construct for South Africa[…]  for cognitive assessment is needed.Where we don't start with an existing model or understanding of intelligence[…]  or how it can be measured.'(P6, Female, Research Psychologist)    Participant 8 suggested: '"First ask yourself what you mean by intelligence […] [what do] IQ-tests measure […] there's [sic] so many different theories about what IQ is […]"; however, "[…] you can't only measure a few cognitive functions and generalise that to [the functioning of] the whole child […] one's got to go back to roots of 'what are you really trying to measure"?' (P8, Female, Counselling and Neuropsychologist)Participant 6 explained that it is difficult for South African stakeholders to develop renewed and personalised psychological constructs, because of their inability to move forward from past negative experiences: '[…] an international psychology conference linked to the Cross-Cultural Conference […] there was a new body called 'Forum for African Psychology' […] I attended to be part of renewal and thinking differently …' (P6, Female, Research Psychologist) She continued: '"[…] my experience was disappointing […] there was a lot of criticism against what was before and I quite accept that many things were not done with everybody in mind …" however, "[…] it's easy to say what we never want to see again, but it is not so easy to say: 'Well, then what should we do"?' (P6, Female, Research Psychologist) Despite this, participants still pursued the idea of creating a unique framework, suggesting the representative input of various groups in South Africa.Sub-theme 4.2: We need input from all different groups represented in South Africa to really understand uniquely how we can define and conceptualise intelligence Stating this sub-theme clearly, Participant 6 suggested: '[…] we need to involve people of all cultural and language groups in this journey […] We need input from all different groups represented in South Africa […].' (P6, Female, Research Psychologist) Elaborating her point: '[…] to really understand uniquely how we can define and conceptualise intelligence […] [that is] unique for South Africa, related to the concept of Ubuntu […] linked to their cultural way of being […].' (P6, Female, Research Psychologist) Providing practical examples: '[…] questions could be revised and be made more contextapplicable […] more in trend with the child's life today.'(P10, Female, Educational Psychologist) '[…] the context is very different now […] questions really need to be revised […] contextualised and generalised for all the settings within the country.' (P5, Female, Educational Psychologist) Adaptation suggestions: '[…] make it more interesting, more creative […] those pictures and illustrations are really quite old-fashioned and come from "long-gone".' (P10, Female, Educational Psychologist) '"[…] scenarios […] [should] not be specific for a certain setting only […] [as] a child that is not exposed to certain settings […]" for instance, "[…] an outdated question about a postage and a stamp […] today's child don't [sic] relate to that.I would rather they be generalised for situations [which] children are now exposed [to]".' (P5, Female, Educational Psychologist) Participant 2 mentioned updating norms, among other updates: '[…] a few updates […] even on norms in South Africa, 'cause […] our culture and demographics have changed quite considerably.'(P2, Female, Psychometrist)

TABLE 1 :
Demographic information of participants.

' number Race Gender Field/type of HPCSA registration Years of practice as registered practitioner Administered intelligence tests to South African school learners Adapted and/or developed intelligence test(s)
HPCSA, Health Professions Council of South Africa.

TABLE 2 :
Demographic information of learners assessed.

TABLE 3 :
Questions asked during semi-structured interviews.Do you believe the locally administered intelligence tests are applicable to South African school learners from different ethnic, socio-economic and/or academic backgrounds?Kindly motivate your answer.2. When and for which purpose do you use intelligence tests? 3. Which intelligence test instruments do you use?What is your motivation for selecting this intelligence test instrument(s) above others available in South Africa? 4. Would you recommend any amendments or changes to be made to intelligence tests?If so, which amendments or changes would you recommend?If not, kindly motivate your answer.5.In which way is the context in which you administer intellectual assessments unique from other contexts?6. Were there elements in the assessed learners' contexts that seemed to have acted as barriers or challenges to their test performance?Kindly elaborate.7. Do you think that the intelligence assessment instrument content considers the school learners' strengths internally and externally, or is it deficient-focussed?8. How would you briefly describe your overall experience of administering intellectual assessments on school learners in this unique assessment context?Do you think the administered intelligence tests are applicable to South African school learners from different ethnic, socio-economic and/or academic backgrounds?Kindly motivate your answer.2. When and for which purpose did you develop/adapt an intelligence test(s)? 3. What were the most significant challenges during the development/adaptation of the intelligence test(s)? 4. What were the most significant insights gained during the development/adaptation of the intelligence test(s)? 5. How would you briefly describe your overall experience of developing/adapting intelligence tests for school learners?

TABLE 4 :
Themes and sub-themes.Many schools within lower socio-economic status contexts in South Africa provide substandard education for school learners 2.2 Should therefore be careful how we interpret results of current intelligence measures Need to establish optimal contexts and practices during preparation, assessment, and reporting processes '[…] not relevant anymore […] the pictures and vocabulary […] relevancy is questionable.It's old-school.' (P12, Female, Educational Psychologist) Participant 5 mentioned how the change of context over time has caused incomprehensiveness of test questions: Your Zulu, your Setswana, your Sepedi […] the words in those tests are outdated […].' (P4, Female, Psychometrist) Participant 7 also noted outdated vocabulary use and test items, arguing that if children were unable to answer these questions, it would not necessarily indicate lower cognitive functioning. ' and if you look at the educational component […].' (P6, Female, Research Psychologist) ] to really pull representative test samples and get access to it […] to get thousands of profiles means a thousand tests that need to be administered for an hour and a half long individually.'(P11, Male, Clinical and Industrial Psychologist) "[…] the variety of people and the context […] when you do the norms, you include almost everybody […]" and the "[…] experience of adapting those tests helped me understand the deeper the importance of norms […] it's not something that we should take very lightly".'(P5, Female, Educational Psychologist) '