ARTICLES

 

A retrospective statistical evaluation of an instrument for discovering parental perspectives on learner discipline

 

 

N.A. Broer; K.B. Mampane; N.T. Mollo; J.L. van der Walt; C.C. Wolhuter

Research Unit: Education and Human Rights in Diversity (Edu-HRight), North-West University, Potchefstroom, South Africa. n.a.broer@live.com

 

 

---

ABSTRACT

In this article we report on a theoretical-statistical investigation into the validity and reliability of a self-developed questionnaire for discovering the perspectives of parents in 2 South African provinces regarding the discipline of their children, both at home and in school. The questionnaire was developed because of a paucity of information concerning this perspective on child or learner discipline. The statistical procedure to which the questionnaire was subjected revealed that the validity of all the sections of the questionnaire was satisfactory, but not the reliability of most parts thereof. Only 1 reliable subscale emerged in the process. Despite this, the information gleaned by means of the individual items of the questionnaire remains useful for understanding the parents' perspectives on learner discipline.

Keywords: critical assessment of a measuring instrument; learner discipline; parental perspectives; reliability analysis; validity analysis

---

 

 

Introduction

We report on one specific part of the outcomes of a research project entitled "Perspectives of parents regarding their role in the disciplining of learners." The project initially required a delineation of key concepts and theory development. This was followed by an empirical investigation into parental perspectives on learner discipline, based on the conceptual-theoretical insights that had been developed. For this purpose, a self-constructed questionnaire (see Appendix A) was developed and disseminated among the parents who formed the research sample. In our discussion we concentrate on our retrospective evaluation of the questionnaire by means of statistical procedures and hence does not attend to the substantial outcomes produced by the questionnaire. This will be done in separate articles in due course.

Although the questionnaire under evaluation was constructed for application in the South African context, it is usable in other national contexts as well. The evaluation thereof should be useful to researchers in such other (national, social) contexts once redacted for their own purposes. A valid and reliable instrument to measure what parents think about learner discipline at home and in school is helpful. However, it is necessary, particularly in countries where the relationship between parents and their child's school is under pressure, such as in countries with emerging economies. Knowledge about this issue is required to improve the relationship between parents and schools about learner discipline and ultimately to improve mature learners' social and economic situation.

This article has a dual purpose. It firstly gives a brief sketch of the background leading to the authors' construction of a questionnaire for the purpose of discovering the perspectives of South African parents regarding the discipline of their children, at home and in school. The second purpose is to report on the outcome of a retrospective critical evaluation of the instrument that we subsequently performed. We employed validity analysis and reliability analysis for this purpose. The structure of this article reflects the procedures that were followed in pursuing these aims.

 

Literature Review

The preamble of the South African Schools Act 84 of 1996 (hereafter referred to as the Schools Act, Republic of South Africa [RSA], 1996a) upholds the rights of parents to be involved in the education of their children in schools. Disciplining children is one of the roles that parents should play for their children to be educated and to learn in a disciplined school environment. Section 8(1) of the Schools Act stipulates that parents should be consulted when a code of conduct for learners is adopted. Section 8(6) of the Act also requires that parents be given an opportunity to accompany their children when they attend disciplinary proceedings. This provision ensures that parents participate in the disciplinary process of their children. Section 9(4) of the Act furthermore allows the parents of learners who have been expelled, should they not be satisfied with the disciplinary process, to appeal against the decision of the head of the department through the member of the executive council within 14 days of receiving the notice of expulsion. These legal provisions ensure that parents have a voice in the creation of a disciplined school environment. The Schools Act (Section 8[1], Section 8[6] and Section 9[4]) underscores the importance of parental involvement in the creation of a disciplined school environment. Based on the extant literature, this means that South Africa is a country in which parental involvement in educational and disciplinary matters is promoted.

Learner ill-discipline (unruly or socially unacceptable behaviour) and misconduct (behaviour that borders on criminality, such as bringing a deadly weapon to school) continue to be a serious problem in South African public schools (Monare, 2013:49; Zulu & Wolhuter, 2013:1). Hence, there is a need for parents to assist in ensuring proper discipline in schools.

When parents discipline their children, they are required to do so in terms of the Constitution of RSA, 1996 (hereafter referred to as the Constitution, RSA, 1996b) (Freedom of Religion South Africa v Minister of Justice and Constitutional Development and Others, 2019). Therefore, parents also need to have an understanding of the Constitution (Chapter 2 - Bill of Rights), which will help them apply the provisions of the Constitution without infringing on their children's rights. Parents also have duties and responsibilities regarding the inculcation of a proper attitude and good moral behaviour in their children.

Parental involvement in schools has been an important research topic for a long time. This statement is supported by Dahie, Mohamed and Mohamed (2018:1) who state that "over 30 years of research shows that one of the most effective ways to increase learner achievement is for parents to be actively involved in the education of their children." Several South African studies emphasise the importance of parental involvement in their children's education (Durisic & Bunijevac, 2017:140; Maluleke, 2014:61; Meier & Lemmer, 2015:1; Page, 2016:133-134; Sapungan & Sapungan, 2014:45). Some of the studies have revealed that the majority of parents do participate in the activities that take place at their children's schools (Ntekane, 2018:3; Poole, 2017:67). Other studies have revealed that parents do not participate optimally in school activities because of a variety of factors (Ntekane, 2018:3; St. George, 2010:36-37). Very few studies have been conducted about the perspectives of parents on their role in the disciplining of learners at home and at school (Barge & Loges, 2003:144; Meier & Lemmer, 2015). A study based on parental perceptions regarding the prevailing disciplinary policy was conducted in New York City (Roch, Elsayed & Edwards, 2018).

What is lacking to a degree in the literature are parents' perspectives regarding the state of their own children's discipline at home and in school. Researchers also have not paid sufficient attention to the question of how parents perceive their own role in the promotion of discipline in schools. The literature review underlying the composition of the questionnaire discussed here indicated a gap in extant empirical studies on parents' perspectives on learner discipline in South African schools. Manamela (2015), Mncube (2009), Moloi (2002) and Postma (1986), conducted longitudinal studies that partially addressed the problem of learner discipline from a parental perspective. To fill this research gap, to contribute to the body of knowledge and to proffer strategies to enhance parents' involvement in their children's discipline, it was vital to find ways to determine parental perspectives on children's discipline. For this reason, we constructed an instrument in the form of a questionnaire. We subjected this self-constructed instrument to various statistical tests to acquire an answer to the following research question: Can the instrument that we have constructed be regarded as valid and reliable? In the following sections we report on how answers to this question were found.

 

Methodology

Research Methodology

A questionnaire was employed in this research project to generate both quantitative and qualitative data. This mixed-method approach was designed to provide a valid and reliable depiction of parental perspectives on learner discipline. The statistical analysis presented below was aimed at establishing the psychometric properties of the self-constructed, quantitative section of the questionnaire.

Instrument

A self-constructed questionnaire was developed to determine South African parents' perspectives on the discipline of their children, both at home and in school. The instrument comprised both Likert-type scale items and open-ended qualitative questions. The Likert scale used in the questionnaire included the response options: "I disagree", "I partly disagree", "I partly agree", and "I agree." A mixed-methods approach was employed by incorporating open-ended questions alongside the quantitative items, thereby facilitating a more comprehensive exploration of the subject under investigation. The items included in the questionnaire were developed based on a previously developed conceptual and theoretical framework, which specifically addressed learner discipline, both at home and in the school environment. The questionnaire was made available to the respondents in their respective home languages in order to ensure maximum understanding of the items.

Data Collection

This study was conducted in two South African provinces: North-West and Mpumalanga. A purposive sampling method, a form of non-probability sampling, was employed to choose research sites and participants. This approach was deemed necessary due to the limitations posed by the then prevailing Coronavirus disease (COVID-19) pandemic, which made the use of (stratified) random sampling impossible. The selection of specific schools was based on their availability during the pandemic. The research sites comprised various school types, including former Model C schools, township schools, and rural schools. Questionnaires were distributed to these schools for parents, using methods that minimised direct human contact in order to adhere to pandemic safety protocols. The completed questionnaires were retrieved from the schools on predetermined dates. A sample of the parent population was selected from each school. A total of 16 schools participated, yielding responses from 448 parents. A limitation of this sample was that the results could not be generalised to the entire country. However, for purposes of this article, this is not a constraint, as the generalisation of the findings was not the objective of this study.

Data Processing

Specialised survey analysis software was used to analyse the data. The Statistical Package for the Social Sciences (SPSS) was employed for purposes of quantitative data analysis, and both validity and reliability analyses were conducted. Several key decisions were taken in the data processing stage, such as how to treat cells that had been left empty by respondents. A significant number of cells in the database had been left empty due to respondents offering no answer/response. In such cases, it was decided that questionnaires missing responses to more than 10% of the 90 items (excluding the biographical data) would be excluded from the final dataset. This decision resulted in 402 of the original 448 questionnaires being retained in the dataset. In seventy (N = 70) cases, respondents did not adhere to the guidelines provided for the completing some of the items. These erroneous responses were also removed from the dataset.

Data Analysis

Exploratory factor analysis was applied to the four sections of the questionnaire to establish the validity thereof. These four sections covered the following sub-issues addressed in the questionnaire: Parents' or the respondents' perspectives regarding

1) the state, causes and effects of learner discipline problems in their children's schools;

2) the state of discipline of their children at home;

3) the management of learner discipline in schools, and

4) their role in curbing the discipline problems of their children in school.

Cronbach's alphas were computed to establish the reliability (stability) of the extracted factors.

Evaluation Criteria

To achieve credibility for the entire project, measures were taken to ensure that all the critical aspects of research, namely, the theoretical framework, the construction of the questionnaire, and the data collection and processing methods would meet standard academic criteria (Nieuwenhuis, 2016:125). Factor analysis was applied to ensure that the questionnaire measured what it was intended to measure. Cronbach's alphas were calculated (Leedy & Ormrod, 2013:91) to determine the reliability of the questionnaire. These calculations were only done only in Sections 1 to 3 of the questionnaire because they produced quantitative data. Factor analysis was done separately for each section. Although Section 3 also contained quantitative data, factor analysis was not considered applicable because of the dichotomous (Yes/No) character of the response scale. The Kaiser-Meyer-Olkin test minimum of 0.50 was applied to determine whether the data were suited for factor analysis. Other generally accepted criteria for the evaluation of data were also applied, such as the communalities minimum value of 0.25, a factor loading minimum of 0.40 and an alpha minimum score of 0.70.

 

Results

Validity and Reliability of Sections 1 to 3

The Kaiser-Meyer-Olkin test applied to the data of these sections yielded values higher than 0.50 in all cases (see Table 1). They were, therefore, considered to be suitable for confirmatory factor analysis (CFA). Cronbach's alphas were calculated for each section to determine its reliability. The results of all these analyses are presented in Table 1.

 

 

Section 1, containing five statements, was about the state, causes and effects of learner discipline problems at schools. The CFA applied to the five items in Section 1 revealed two factors relevant to all five items. The communalities minimum and factor loading minimum were sufficient, but not the alpha scores.

Section 2, containing 13 statements, was about the state of discipline among children at home. The CFA yielded four factors. The communalities and the factor loading minima were also met. It emerged that there was an imbalance in the number of items per factor - the fourth factor contained only one item. For this reason, an alpha score could not be calculated for the fourth factor. To overcome this, the forced one-factor solution described in the next subsection was run.

Section 3, containing six items, dealt with the management of learner discipline in schools. The CFA applied to the data in this section yielded two factors. The communalities and factor loading minima were adequate, but the alpha scores were too low in terms of the set criteria.

The Forced One-factor Solution for Section 2

As mentioned, a forced one-factor solution had to be run for Section 2, and items with a communality score lower than 0.25 were deleted. The alpha score for that factor was satisfactory (see Table 1). This yielded a single factor consisting of eight items. A factor analysis after adapting the original number of items in the section resulted in a scale that reflected the state of discipline among children at home and met the criteria for validity and reliability. Tables 2 and 3 contain the results of the factor analysis applied to the data of Section 2. The table reflects the eight components (factors) that, among them, explained the variance with respect to the state of discipline among children at home. Table 3, in turn, explains which items in the questionnaire loaded high onto Component 1 (Factor 1) and hence explained most of the variance, namely 38%.

 

 

 

 

Discussion

As mentioned, the purpose with this article was to answer the question of whether the self-constructed instrument that we used was valid and reliable. Our computations yielded a four-fold answer.

Sub-answer 1

Five items in the questionnaire (Section 1) were intended to discover the respondents' (parents') perspectives regarding the state, causes and effects of learner discipline problems in their children' s schools. It transpired that these items did not form a single scale because only two factors had surfaced. This can be ascribed to the fact that two of the statements in the questionnaire had been formulated with linkages to previous statements (e.g., "If your answer to item x was do you ..."). Only some respondents reacted suitably to these complex statements, and this led to the instability or unreliability of this section. This section, therefore, could be deemed to have failed in reliably discovering the parents' perspectives regarding the state, causes and effects of learner discipline in their children's schools.

Sub-answer 2

The 13 items in Section 2 of the questionnaire were formulated for the purpose of determining the parents' perspectives on the state of the discipline of their children at home. The first application of CFA to these statements did not yield a satisfactory resolution in that the items were unevenly spread over the four extracted factors. An additional problem in this case was that the factors were too heterogeneous to be suitably labelled. However, the subsequent principal component analysis with forced rotation of one factor yielded a reliable scale of eight items. This scale proved to be appropriate for discovering the parents' perspectives on the issues mentioned above (see Tables 2 and 3).

Sub-answer 3

The next task was to determine the parents' perspectives regarding the management of learner discipline in schools. Six statements in the section yielded two factors that proved to be difficult to label due to their heterogeneous composition. The alpha scores of both were also too low. Section 3 was found not to be an appropriate instrument for determining the parents' perspectives regarding the management of learner discipline in schools.

Sub-answer 4

The last section of the questionnaire was about discovering the parents' perspectives regarding their role in curbing the discipline problems of their children in school. No psychometrical analysis could be applied in this case because of the dichotomous character of the response scale. The difficulties in this section entailed a) the limited number of items, b) the interdependence among the items and c) the possibility of respondents choosing "not applicable" for one of the items. Section 3 was, therefore, found not to be an appropriate instrument for determining the parents' perspectives on their role in the curbing of discipline problems of their children in school.

 

Recommendations

Based on the study reported on here, the following points are recommended for the purpose of avoiding the abovementioned pitfalls and shortcomings:

• To construct a valid and reliable instrument, much more than five items should have been included in Section 2. This ought to enable the retainment of a sufficient number of items after factor analysis.

• The practice of respondents having to respond to a linkage between two items should be avoided. This practice leads to confusion and, consequently, to unusable data.

• In order to be in line with the other sections of the questionnaire, thereby being amenable to psychometric analysis, it would have been better to format Section 3 in the form of a 4-point Likert scale instead of a dichotomy.

 

Conclusion

In this article we report on attempts to develop a valid and reliable instrument to measure parents' thoughts about learner discipline at home and in school. The authors hope this will serve researchers, especially those who work in countries where the relationship between parents and schools is under pressure. Solid research outcomes can contribute to improving education in countries with emerging economies and socially deprived population groups.

Although three of the four sections of the instrument that we had constructed for the purpose of determining various parental perspectives regarding learner discipline at home and at school were found to be lacking in terms of the tests that we used to determine its validity and reliability, the instrument cannot be deemed to be unusable. Not only was a part of one of its sections found to be valid and reliable or stable, but all the data yielded by the individual items in the other three sections remained serviceable in that they cast light on parental perspectives on learner discipline at home and in the schools attended by the respondents' children. The data thus offered usable information despite the fact that the different sections did not form coherent scales.

The following are examples of such information. Forty per cent of the respondents disagreed (fully or partially) with the statement that the standard of discipline in schools in South Africa was very good; only 19% disagreed (fully or partially) with the statement, "The discipline in my child's school is very good." Furthermore, 82% of the respondents were of the opinion that their children's schools sufficiently involved parents in learner discipline matters. Nearly half of the respondents (46%) declared that they had participated in a disciplinary proceeding regarding the behaviour of their children, and 95% of this group declared that their participation in their children's disciplinary procedures had improved their children's behaviour at school.

 

Authors' Contributions

NAB supervised the writing of the article and contributed to the statistical and methodological sections. KBM, NTM, JLvdW and CCW contributed to the other sections. JLvdW revised the initial draft version of the manuscript.

 

Notes

i. Published under a Creative Commons Attribution Licence.

 

References

Barge JK & Loges WE 2003. Parent, student, and teacher perceptions of parental involvement. Journal of Applied Communication Research, 31(2):140-163. https://doi.org/10.1080/0090988032000064597        [ Links ]

Dahie AM, Mohamed AA & Mohamed RA 2018. The role of parental involvement in student academic achievement: Empirical study from secondary schools in Mogadishu-Somalia. International Research Journal of Human Resources and Social Sciences, 5(7):1-24. Available at https://www.researchgate.net/profile/Abdulkadir-Dahie/publication/326957190_THE_ROLE_OF_PARENTAL_INVOLVEMENT_IN_STUDENT_ACADEMIC_ACHIEVEMENT_EMPIRICAL_STUDY_FROM_SECONDARY_SCHOOLS_IN_MOGADISHU-SOMALIA/links/5b6d9f4945851546c9fa298f/THE-ROLE-OF-PARENTAL-INVOLVEMENT-IN-STUDENT-ACADEMIC-ACHIEVEMENT-EMPIRICAL-STUDY-FROM-SECONDARY-SCHOOLS-IN-MOGADISHU-SOMALIA.pdf. Accessed 10 August 2025.         [ Links ]

Durišić M & Bunijevac M 2017. Parental involvement as a important factor for successful education. Center for Educational Policy Studies Journal, 7(3): 137-153. https://doi.org/10.26529/cepsj.291        [ Links ]

Leedy PD & Ormrod JE 2013. Practical research: Planning and design (10th ed). Upper Saddle River, NJ: Pearson Education.         [ Links ]

Maluleke SG 2014. Parental involvement in their children's education in the Vhembe District: Limpopo. MEd dissertation. Pretoria, South Africa: University of South Africa. Available at https://www.proquest.com/docview/3224600007?porigsite=gscholar&fromopenview=true&sourcetype=Dissertations%20&%20Theses. Accessed 9 August 2025.         [ Links ]

Manamela LM 2015. The role of parental involvement in improving discipline in secondary schools of the Kgakotlou circuit. MEd dissertation. Pretoria, South Africa: University of South Africa. Available at https://www.proquest.com/docview/3224571122?pq-origsite=gscholar&fromopenview=true&sourcetype=Dissertations%20&%20Theses. Accessed 11 August 2025.         [ Links ]

Meier C & Lemmer E 2015. What do parents really want? Parents' perceptions of their children's schooling. South African Journal of Education, 35(2):Art. # 1073, 11 pages. https://doi.org/10.15700/saje.v35n2a1073        [ Links ]

Mncube V 2009. The perception of parents of their role in the democratic governance of schools in South Africa: Are they on board? South African Journal of Education, 29(1):83-103. https://doi.org/10.15700/saje.v29n1a231        [ Links ]

Moloi PIM 2002. The role of the parents in discipline as an aspect of school safety. MEd dissertation. Johannesburg, South Africa: Rand Afrikaans University. Available at https://www.proquest.com/docview/2572687975?pq-origsite=gscholar&fromopenview=true&sourcetype=Dissertations%20&%20Theses. Accessed 9 August 2025.         [ Links ]

Monare LM 2013. The management of learner discipline by female primary school principals in the Bojanala Region of North-West Province. MEd dissertation. Mafikeng, South Africa: North-West University. Available at https://repository.nwu.ac.za/items/297511b7-16d5-4ca7-b9eb-fa88f5997785. Accessed 9 August 2025.         [ Links ]

Nieuwenhuis J 2016. Analysing qualitative data. In K Maree (ed). First steps in research (2nd ed). Pretoria, South Africa: Van Schaik.         [ Links ]

Ntekane A 2018. Parental involvement in education. Potchefstroom, South Africa: North-West University. https://doi.org/10.13140/RG.2.2.36330.21440        [ Links ]

Page RD 2016. The role that parents play in their children's academic progress at a previously disadvantaged primary school in Cape Town. MEd thesis. Bellville, South Africa: University of the Western Cape. Available at https://uwcscholar.uwc.ac.za/items/cf346529-4045-412a-af63-4cc822387528. Accessed 9 August 2025.         [ Links ]

Poole A 2017. Exploring parental involvement in the educational support of their child with learning difficulties in a low income community. MEd thesis. Stellenbosch, South Africa: Stellenbosch University. Available at https://core.ac.uk/reader/188222070. Accessed 9 August 2025.         [ Links ]

Postma W 1986. 'n Fundamenteel-opvoekundige presisering van oueronderwyserkontak. DEd thesis. Potchefstroom, South Africa: Potchefstroom University for Christian Higher Education.         [ Links ]

Republic of South Africa 1996a. South African School Act, 1996 (Act No. 84, 1996). Government Gazette, 377(17579):1-50, November 15.         [ Links ]

Republic of South Africa 1996b. The Constitution of the Republic of South Africa (Act 108 of 1996). Pretoria: Government Printer.         [ Links ]

Roch CH, Elsayed MAA & Edwards J 2018. Students' and parents' perceptions of disciplinary policy: Does symbolic representation matter? The American Review of Public Administration, 48(4):329-345. https://doi.org/10.1177/0275074016686420        [ Links ]

Sapungan GM & Sapungan RM 2014. Parental involvement in child's education: Importance, barriers and benefits. Asian Journal of Management Sciences & Education, 3(2):42-18.         [ Links ]

St. George CY 2010. How can elementary teachers collaborate more effectively with parents to support student literacy learning? Delta Kappa Gamma Bulletin, 76(2):32-38.         [ Links ]

Zulu CB & Wolhuter CC 2013. Preface to special edition on learner discipline problems in schools [Special edition]. Koers - Bulletin for Christian Scholarship, 78(3):Art. #2153, 2 pages. https://doi.org/10.4102/koers.v78i3.2153        [ Links ]

Legal Authorities

Freedom of Religion South Africa v Minister of Justice and Constitutional Development and Others (CCT320/17) [2019] ZACC 34; 2019 (11) BCLR 1321 (CC); 2020 (1) SA 1 (CC); 2020 (1) SACR 113 (CC) (18 September 2019).

 

 

Received: 2 October 2023
Revised: 16 October 2024
Accepted: 31 March 2025
Published: 31 August 2025

 

 

Appendix A