2. Materials and Methods

2.1 Study Setting

A questionnaire-based cross-sectional survey was conducted in Visakhapatnam (GVMC –Greater Visakhapatnam Municipal Corporation), a metropolitan city (with a population of more than 1 million people) in Andhra Pradesh, India (see Figure 2). Visakhapatnam is the first largest city and a proposed executive capital city of the newly bifurcated state of Andhra Pradesh. The city has a population of 1.7 million, a core urban area of 122 km2 and a total metropolitan area of 539.95 km2; the city is administratively organised into six zones and 72 wards. Further, the city consists of major ports, industries, and educational institutions in the state, attracting many migrants from the neighbouring regions.

$C:\Users\Admin\Desktop\rrk\full map.jpg$

Figure 2. Map representing the location of Visakhapatnam city in Andhra Pradesh, India.

2.2 Participants

The targeted respondents were the parents of elementary school pupils (5-14 years old). Two sampling methods were used to collect the data and received 247 samples from a non-probabilistic snowball sampling technique (circulating questionnaire link/QR code through SMS and WhatsApp) and 220 samples from a simple random sampling (Circulating questionnaire form at the neighbourhood parks) during April 2021 to August 2021. Out of 467 observations, 409 valid responses are retained for the analysis. The purpose of the study was communicated, and informed consent was obtained from the participants.

2.3 Materials, Equipment, Procedures and Protocols

Indicators identified from previous studies to formulate the questionnaire and discussed with subject experts, i.e., school teachers and school-going children's parents, to contextualize the instrument. The questionnaire involves demographic, socio-economic, neighbourhood environment, and spatial and non-spatial parameters of schools. Microsoft Forms (an online survey tool) is used to prepare a questionnaire form for distribution to the targeted respondents. Each question in the questionnaire form was asked in two languages, English and Telugu (a regional language in the study context), for the convenience of the participant. The author has sufficient knowledge of both languages. Hence, the translation of the questionnaire reflects the same meaning in both languages. The data types collected are nominal, ordinal, and interval/ratio with a structured closed-ended and open-ended questionnaire. In this paper, only a selected closed-ended questionnaire is taken into consideration.

2.4 Data Analysis

Descriptive statistics were used to represent the characteristics of the overall sample in frequencies and percentages. Variables like home-to-school distance, commute mode, and school availability in the neighbourhood have been cross-tabulated with frequencies. It has provided information about the proportion of children studying within the neighbourhood and commuting outside with varied distances and commute modes. The Chi-square test is a distribution-free non-parametric test; when the Level of measurement of variables is nominal/ordinal, the sample size of the study groups is unequal, and the distribution is non-normal, this test gives robust results and detailed information about the interdependency of the variables (Mchugh, 2013). The variables in our study satisfy the (Mchugh, 2013) chi-square test criteria and do not require preliminary tests to check the normality. Hence, further analysis was conducted in non-parametric tests using Pearson's chi-square (χ2) test of independence to investigate the significant relations between the dependent and independent variables. After obtaining the significant variables, a multinomial logistic regression analysis was conducted using the forward method with the significant variables to predict the odds ratios of the home-to-school distance. All the statistical tests were conducted using IBM® SPSS® Statistics Version 27.

3. Results

3.1 Children studying within the neighbourhood vs outside

As shown in Table 2, out of 409 parents' responses, 362 parents reported school availability within the range of 1km from their home, and the remaining 47 parents reported unavailability or unaware of the school in the 1km vicinity. However, out of 362 parents, only 136 parents (i.e., 59 parents with single children, 69 parents having two children, and 8 parents having three children) reported that their children are studying at the same school, and 193 parents (i.e., 144 parents with single children, 45 parents having two children, and 4 parents having three children) reported that they are sending their children outside the neighbourhood for school. Surprisingly, 33 parents said at least one of their children is at the neighbourhood school, and the rest are studying outside the neighbourhood.

Table 1: Characteristics of the sample.

Characteristics of the sample			Characteristics of the sample
Variable	Frequency	Percentage	Variable	Frequency	Percentage
Number of Respondents	409	100%	Board of Child School
Gender of the Child			Local / State govt. board	297	72.60%
Male	230	56.20%	CBSE	82	20.00%
Female	179	43.80%	ICSE	30	7.40%
Level of Education of the Child			Medium of instruction
Primary	208	50.90%	Regional language	48	11.70%
High School	201	49.10%	English	361	88.30%
Number of Children in the Home			Type of management of the school
1	232	56.70%	Public	108	26.40%
2	160	39.10%	Semi-public	60	14.70%
3	17	4.20%	Private	241	58.90%
Education of the parent			Level of education of the school
Did not attend the school	22	5.40%	Up to Primary	83	20.30%
Up to matriculation	46	11.20%	Up to Upper-Primary	70	17.10%
Graduate	194	47.40%	Up to Secondary (High School)	256	62.60%
Above graduation	147	35.90%	Type of education of the school
Occupation of the Parent			Co-Education	380	92.90%
Both are not working	26	6.40%	Only Girls	20	4.90%
Single parent working	290	70.90%	Only Boys	9	2.20%
Both are working	93	22.70%	Teaching Methods
Annual income of the family (Indian Rupees)			Traditional	140	34.20%
< 3,00,000	101	24.70%	Modern / Digital	36	8.80%
3,00,001 - 6,00,000	133	34.00%	Combined	233	57.00%
6,00,001 - 12,00,000	123	30.10%
> 12,00,000	46	11.20%

Table 2: Children studying within the neighbourhood vs outside.

		Is the school within 1km of your residence available?		If yes, is your child studying at the same school?
		Yes	No	Yes, all my children study at the same school	Yes, one of my child studying at the same school	No
Number of School Children in the Family	1	204	28	59 (59)	1(1)	144 (144)
	2	143	17	69 (138)	29 (29)	45 (90)
	3	15	2	8 (24)	3 (3)	4 (12)
	Total	362	47	136 (221)	33 (33)	193 46)

3.2 Mode of the school commute and home-to-school distance

As shown in Table 3, children travel to school by active modes (walking and bicycling) up to the range of 2km from home; as the distance increases, the mode of commute changes to the motorised vehicle. However, the commute mode may vary according to the child's gender and level of education. The Pearson correlation results indicate a significant relationship between the variables of the child's school commute mode and the distance of the school from home (χ2 = 250.078, df = 15, p-value < 0.001).

Table 3: Mode of school commute vs home-to-school distance.

Mode of School Commute	Distance of the child's school from home
Mode of School Commute	< 1 km	1-2 km	2-5 km	> 5 km	Total
Walking	45	13	0	0	58
Bicycling	25	2	2	0	29
Rickshaw/Auto	47	45	22	13	127
Dropping/Pick-up by parent or caretaker	11	21	27	6	65
School van/bus	7	13	45	51	116
Public transport	0	0	2	12	14

3.3 Correlation among variables (Refer to Table 4)

3.3.1 Demographic factors vs home-to-school distance

Gender

Our sample of responses represents 43.8% girls and 56.2% boys school admissions. Regarding the distance of the school from home, girls were slightly lower compared to boys in the school admissions, whereas, while distance increased from home to school, boys' numbers slightly increased. However, the difference was not statistically significant (χ2 = 0.646, df = 3, p-value = 0.886). (Easton & Ferrari, 2015) reported that gender moderated with age could significantly affect the commute distance as a girl's age may increase commute distance.

Age / Level of the child's education

Parents of primary school children were more likely to send their children to schools near home (39.4%, 23.6%, 20.2%, and 16.8% in <1km, 1-2km, 2-5km, and above 5km, respectively). Whereas in high-school children, the number of children was significantly increasing in comparison to primary-level children as the distance increased from the home (26.4%, 22.4%, 27.9%, and 23.4%, in <1km, 1-2km, 2-5km, above 5km, respectively). There is a significant difference in the Level of the child's education and the distance of the school from home (χ2 = 10.039, df = 3, p-value < 0.05). The results correlate with (Easton & Ferrari, 2015) findings, where a child's age/level of education is very consistent and significant in association with home-to-school distance.

Number of children in the family

Regarding the number of school-going children in the family, our sample reported 232 (56.7%) families having a single child, 160 (39.1%) families having two children and the rest 17 (4.2%) families having three children. Our results describe that families with two or more children prefer schools within 2km from home, whereas single-child parents' preferences vary regardless of the distance. Our findings report a significant difference in the number of school children in the family and home-to-school distance (χ2 = 21.040, df = 6, p-value < 0.05).

3.3.2 Socio-economic factors vs home-to-school distance

Parent's educational qualification

Of all the 409 observations, 174 (35.9%) respondents are above graduation; among them, 37 parents (25.2%) are sending their children to the school in the neighbourhood, another 105 parents (71.4%), including the above 37 respondents' children are studying at the school within 5km from home, and 42 parents (28.6%) are sending their children to the school above 5km from home. Similarly, 194 (47.4%) respondents are graduates; among them, 72 parents (37.1%) are sending their children to the school in the neighbourhood, 166 parents (85.4%), including the above 72 respondents' children are studying at the school within 5km from home, and 28 parents (14.4%) are sending their children to the school above 5km from home. Likewise, 46 (11.2%) respondents are up to matriculation; among them, 17 parents (37%) are sending their children to the school in the neighbourhood, and the rest 29 parents (63%) are sending their children to the school outside the neighbourhood. Lastly, 22 (5.4%) respondents did not attend schooling; among them, 9 parents (40.9%) are sending their children to the school in the neighbourhood, and the 18 parents (81.8%) children including above 9 are studying within the 5km, and the remaining 4 parents (18.2%) are sending their children above 5km. These results describe a slight relation between an increase in the parent's education level to the Distance of the children's school from home; however, the difference was not statistically significant (χ2 = 14.991, df = 9, p-value = 0.091).

Annual income of the family

Of all the 409 observations, 101 respondents (24.7%) have income below 3,00,000 and send their 45 (44.6%) children to the school in the neighbourhood, 92 (91.1%) children are studying the range of 5km from home, and only 9 (8.9%) children are travelling more than 5km from home for schooling. Similarly, 139 respondents (34%) have an income between 3,00,001-6,00,000 and send their 52 (37.4%) children to the school in the neighbourhood, 115 (82.7%) children are studying the range 5km from home, and 24 (17.3%) children are travelling more than 5km. Likewise, 123 respondents (30.1%) have an income of 6,00,001-12,00,000 and send their 25 (20.3%) children to the school in the neighbourhood, 82 (66.7%) children are studying in the range of 5km from home, and 41 (24.7%) children are travelling more than 5km from home for schooling. And finally, 46 respondents (11.2%) have an income of above 12,00,000; among them, 13 (28.3%) parents send their children to the school in the neighbourhood, 38 (82.6%) are within the range of 5km from home, and the rest 8 (17.4%) of them are travelling more than 5km. Pearson correlation suggests a significant correlation between the annual income of the family and the Distance of the children from home (χ2 = 40.251, df = 9, p-value < 0.001).

Parents occupation

In our sample, 290 (70.9%) respondents reported single parents are working, 93 (22.7%) said both parents are working, and only 26 (6.4%) reported not working anywhere. The correlation between parents' occupation and home-to-school distance did not show any statistical significance (χ2 = 1.290, df = 6, p-value = 0.972).

3.3.3 School's non-spatial factors vs home-to-school distance

Board of the school

Of all the 409 observations, 297 parents (72.6%) are sending their children to the Local / State government board schools; among them, 117 (39.4%) parents send their children to the neighbourhood school, 257 (86.5%) children are studying in the range of 5km from home, and only 40 (13.5%) children are travelling more than 5km from home. Similarly, 82 parents (20.1%) are sending their children to CBSE schools; among them, 13 (15.9%) parents send their children to the neighbourhood school, 52 (63.4%) children are studying in the range of 5km from home, and 30 (36.6%) children are travelling more than 5km from home. Finally, 30 parents (7.3%) are sending their children to the ICSE schools; among them, 5 (16.7%) parents send their children to the neighbourhood school, 18 (60%) children are studying in the range of 5km from home, and 12 (40%) children are travelling more than 5km from home. These results show that most children studying in local/state govt board schools travel less than 5km for schooling from their home, whereas children studying at CBSE and ICSE schools travel more than 5km. Although most children study at local schools, there is a significant difference in the distance of the school from home and the board of children's school (χ2 = 39.256, df = 6, p-value < 0.001).

Our statistical model (see table 5) presents the probability of preference for CBSE schools compared to the local board having 1.198, 4.118, and 6.721 times in 1-2km. 2-5km and above 5km categories, respectively, with the base category of home-to-school distance below 1km. Similarly, the probability of preference for ICSE schools compared to the local board having 1.200, 2.370, and 6.669 times in 1-2km. 2-5km and above 5km categories, respectively, with the base category of home-to-school distance below 1km.

Medium of instruction

The majority of the parents choose English as an instruction language rather than a local/regional language; out of 409 observations, 361 (88.3%) parents sending their children to schools of English language instruction, and only 48 (11.7%) parents are sending to the schools of local/regional language instruction. Though the majority prefer English instruction as a major criterion for their children's school, the distance of the school from home and the medium of instruction does not significantly differ (χ2 = 6.789, df = 3, p-value = 0.079).

School management

Nearly two-thirds of the sample represents private schools over public and semi-public together. Public school students decrease as home-to-school distance increases, whereas private school children remain consistent. However, the home-to-school distance over school management is not statistically significant at 95% CI (χ2 = 12.132, df = 6, p-value = 0.059).

Level of education offered by the school

Our sample represents primary school children with 208 members. However, only 83 (40%) were enrolled in schools offering primary education, and the rest, 125 (60%), were enrolled in schools that offer up to elementary and secondary level. Chi-square statistics show the association between home-to-school distance and the level of education offered by the school in school selection (χ2 = 18.043, df = 6, p-value < 0.05). However, the child's education level needs further examination of this significance.

Type of education offered by the school

In the type of schooling, 93% of the sample represents co-education schools, and only 7% represents only girls and only boys schools together. The chi-square statistic is not significant on home-to-school distance and type of school (χ2 = 3.464, df = 6, p-value = 0.749).

Table 4: Correlation among variables.

Variable	Descriptive statistics (frequency)				Pearson Chi-Square analysis
	Distance of the child's school from home
	< 1 km	1-2 km	2-5 km	> 5 km	χ2	df	Sig.
Gender of the Child					0.686	3	0.886
Male	74	51	58	47
Female	61	43	40	35
Level of Education of the Child					10.039	3	< 0.05 (0.018)
Primary	82	49	42	35
High School	53	45	56	47
Number of Children in the family					21.040	6	< 0.05 (0.002)
1	58	57	70	47
2	69	32	26	33
3	8	5	2	2
Education of the parent					14.991	9	0.091
Did not attend the school	9	3	6	4
Up to matriculation	17	10	11	8
Graduate	72	44	50	28
Above graduation	37	37	31	42
Occupation of the Parent					1.290	6	0.972
Both are not working	9	6	7	4
Single parent working	96	68	70	56
Both are working	30	20	21	22
Annual income of the family					40.251	9	< 0.001 (0.000)
< 3,00,000	45	27	20	9
3,00,001 - 6,00,000	52	25	38	24
6,00,001 - 12,00,000	25	24	33	41
> 12,00,000	13	18	7	8
Board of child school					39.256	6	< 0.001 (0.000)
Local / State govt. board	117	72	68	40
CBSE	13	15	24	30
ICSE	5	7	6	12
Medium of instruction					6.789	3	0.079
Regional language	19	16	7	6
English	116	78	91	76
Type of management of the school					12.132	6	< 0.05 (0.059)
Public	36	32	21	19
Semi-public	16	12	12	20
Private	83	50	65	43
Education offered by the school					18.043	6	< 0.05 (0.006)
Up to Primary	41	11	13	18
Up to Upper-Primary	19	22	18	11
Up to Secondary (High School)	75	61	67	53
Type of education of the school					3.464	6	0.749
Co-Education	128	85	91	76
Only Boys	4	7	4	5
Only Girls	3	2	3	1
Teaching Methods					3.589	6	0.732
Traditional	52	34	32	22
Modern / Digital	11	9	8	8
Combined	72	51	58	52

3.4 Results of Multinomial Logistic Regression Analysis: (Refer to Table 5)

After identifying the significant predictor variables using Pearson Chi-Square correlation, a multinomial logistic regression model was performed to predict the odds of parental preferences on school selection (dependent variable: the distance of the child's school from home 1 = <1km, 2 = 1-2km, 3 = 2-5km, 4 = above 5km. and independent/predictor variables are the number of school children at home, the child's education level, the family's annual income and the board of education of the child's school). The model was performed using the stepwise method of forward entry.

3.4.1 Model Fitting Information

The model fitness was measured using the likelihood ratio (LR) chi-square test, comparing the model's fitness with the complete set of predictors with an intercept-only, or null, model (no predictors). Based on the LR test, the model containing the complete set of predictors represents a significant improvement in fit relative to a null model (LR χ2 (21) = 97.796, P<0.001). This result proves at least one population slope is non-zero. The Pearson (0.933) and Deviance (0.851) statistics prove that the model is fit since the P-value is > 0.05. The Pseudo R-square measures are Cox and Snell (0.289), Nagelkerke (0.310) and Mc Fadden (0.128), which implies a moderately strong model referencing to the odds ratio scale in the results (Bo Hu, et.al., 2006). The model accounts for 20% to 33% of the variance and represents relatively decent-sized effects. The likelihood ratio test provides that the independent or predictor variables, number of school children at home, education level of the child, board of education and family's annual income are significant contributors.

3.5 Predictors of the home-to-school distance

Table 5 shows the predicted statistics of the model concerning the base category as the distance of the school from home is less than 1km. In the 1-2 km category, none of the predictor variables is statistically significant at p < 0.05. Only the board of education CBSE & ICSE positively correlates with the dependent category. In contrast, the remaining predictor variables negatively correlate to the home-to-school distance. Similarly, in the 2-5 km category, board of education CBSE & ICSE, number of children and Level of child education are statistically significant at p < 0.05 except for the family's annual income. Among them, the child's board of education and education level have a positive correlation, and the number of school children has a negative correlation. Finally, in the category above 5km, the board of education CBSE and education level are statistically significant at p < 0.05.

Table 5: Results of Multinomial Logistic Regression.

Variables	B	(S.E)	P - Value	Odds Ratio (Exp . B)		95% C.I
					Lower		Upper
Distance of the school from home = 1-2km
Board of Child's School = Local/State Govt	-	-	-	-	-		-
Board of Child's School = CBSE	0.181	0.499	0.717	1.198	0.451		3.185
Board of Child's School = ICSE	0.182	0.697	0.794	1.200	0.306		4.703
Number of School Children	-0.529	0.237	0.025	0.589	0.370		0.937
Education Level of the Child	0.419	0.279	0.134	1.520	0.879		2.628
Annual income of the family = Below 3,00,000	-	-	-	-	-		-
Annual income of the family = 3,00,001 - 6,00,000	-0.311	0.350	0.374	0.733	0.369		1.454
Annual income of the family = 6,00,001 - 12,00,000	0.412	0.399	0.302	1.510	0.691		3.303
Annual income of the family = Above 12,00,000	0.652	0.560	0.244	1.919	0.641		5.745
Distance of the school from home = 2-5km
Board of Children's School = Local/State Govt	-	-	-	-	-		-
Board of Children's School = CBSE	1.415	0.471	0.003	4.118	1.636		10.366
Board of Children's school = ICSE	0.863	0.726	0.235	2.370	0.571		9.828
Number of School Children	-1.046	0.263	0.000	0.351	0.210		0.589
Education Level of the Child	0.813	0.285	0.004	2.254	1.290		3.938
Annual income of the family = Below 3,00,000	-	-	-	-	-		-
Annual income of the family = 3,00,001 - 6,00,000	0.257	0.358	0.473	1.293	0.641		2.610
Annual income of the family = 6,00,001 - 12,00,000	0.584	0.427	0.171	1.793	0.777		4.140
Annual income of the family = Above 12,00,000	-0.958	0.671	0.154	0.384	0.103		1.431
Distance of the school from home = >5km
Board of Children's School = Local/State Govt	-	-	-	-	-		-
Board of Children's School = CBSE	1.905	0.468	0.000	6.721	2.688		16.804
Board of Child's School = ICSE	1.898	0.671	0.005	6.669	1.789		24.859
Number of School Children	-.461	0.268	0.086	0.631	0.373		1.067
Education Level of the Child	0.829	0.310	0.008	2.290	1.247		4.205
Annual income of the family = Below 3,00,000	-	-	-	-	-		-
Annual income of the family = 3,00,001 - 6,00,000	0.589	0.452	0.193	1.803	0.743		4.376
Annual income of the family = 6,00,001 - 12,00,000	1.303	0.497	0.009	3.681	1.390		9.746
Annual income of the family = Above 12,00,000	-0.517	0.703	0.462	0.596	0.150		2.364

-: Bold figures are significant at p < 0.05

4. Discussion Summary

Many researchers used cross-sectional research design and multinomial logistic analysis to model the school commute distances and mode of commute assessment. In this research, we have used the same methods for investigating parental preferences and analysing how demographic, socio-economic, and school non-spatial attributes are instrumental to home-to-school distance in an urban context in India. Our preliminary results found that many parents send their children to schools outside their neighbourhood, though there is a school facility within their neighbourhood. And it has also shown a significant association between home-to-school distance and the mode of school commute. These findings are similar to previous studies in similar Indian urban contexts (Tetali et al., 2016; N. Singh & Vasudevan, 2018).

As explained in 4.3.1 (Age / Level of the child's education), parents of primary school children prefer schools near home, whereas high school children in our study travelled long distances for schooling. The results correlate with (Easton & Ferrari, 2015) findings, where a child's age/level of education is very consistent and significant in association with home-to-school distance. Regarding a child's gender role, this variable does not directly affect the home-to-school distance in our findings. (Easton & Ferrari, 2015) reported that gender moderated with age could significantly affect the girl's age as the commute distance may increase. We further analysed the influence of the number of children in the family associated with home-to-school distance. In our findings, the number of school-going children in the family negatively correlates with school distance. Families with two or more children send their children to schools within 2km, whereas single-child families' preferences vary with mixed responses. This finding is close to the earlier study by (Ota & Moffatt, 2006) in rural Andhra Pradesh. It talks about opportunities based on family size and birth order, where the younger child in siblings receive privileges over the firstborn and elder children. Such studies were initial and focused on only schooling opportunities, and did not explore any relation to home-to-school distance. recent times, increased access and opportunities in school education have given numerous opportunities to parents in urban areas for various school choices.

Parents' perceived specific non-spatial attributes define the quality of the school. Previous studies focused on school management, medium of instruction, smaller class sizes, teaching methods, and academic attainments are found to be significant factors in parental preferences (Alsuiadi, 2015). In addition to the above factors, the board of education is crucial in school selection in the Indian context, mainly the social construct over the impression of school boards. Usually, neighbourhood schools are local/state board schools that consist of basic infrastructure and facilities and offer application-based learning. In contrast, CBSE and ICSE are magnet schools with higher infrastructure standards, equipped with facilities, laboratories, and libraries, and offer application-based and problem-solving pedagogy. Such schools are mainly located in suburban areas, and very few are in core urban neighbourhoods since they require a more extensive site area. Such schools are accessible to financially resourced families with better parent education levels in urban areas.

5. Conclusion

The findings of this research study highlight that parents' preferences are instrumental in the non-spatial attributes of the school environments. This study explains how the role of demographic, socio-economic factors and school’s non-spatial attributes influences the parents' preferences in school selection concerning home-to-school distance in urban areas. In this regard, the findings from the case of Visakhapatnam show that significant determinants of home-to-school distance are

- Age and the number of school children in the family (demographic),

- family income (socio-economic), and

- attitude toward the board of education and management of the school (non-spatial attributes of the school).

These findings offered a new dimension to rethink the theoretical perspectives that have defined the traditional neighbourhood concepts of Clerence Artur Perry (Aronovici, C. 1939) and (Mumford, 1949), on the importance of school in neighbourhood unit design whilst investigating from a developing country's position. The fact is that cities are ever-changing with their demographic and socio-economic needs. The infrastructure setups for education resources within the urban areas are also drastically changing for the development causes with a competitive spirit. The availability of land resources and affordability have become the open and often latent drivers in the cities. For instance, this study has shown that demographics, socio-economic factors and attitudes toward the school's board of education (non-spatial attributes of the school) have driven parents' preferences in selecting the schools for their children in the case of Visakhapatnam City.

From a methodological position, this study adopted a cross-sectional research design for collecting data and a multinomial logistic model to predict the factors determining the parents' preferences in selecting the schools for their children with respect to home-to-school distance. However, this study has shown a direction for collating and analysing the data on parents' preferences towards school selection. Such models need to be developed from a multi-disciplinary perspective that can assimilate the spatial distribution of such non-spatial parameters in an integrated manner. Nevertheless, this study showed that ongoing and future school revamping schemes are often limited only to the physical and educational aspects of the schools. Instead, it should be a holistic approach scaling from the school setting, creating a sense of neighbourhood unit, and even planning the city transport network.

This study contributes to the existing literature by expanding the theoretical understanding of neighbourhood design, particularly in the context of developing countries. It challenges traditional theories by incorporating the influence of non-spatial attributes of schools and socio-economic factors on parents' preferences, thereby offering a more holistic view of urban educational infrastructure. It highlights the need for policymakers to consider the Integrated Urban Planning and Development policies that integrate educational infrastructure within neighbourhood planning, ensuring accessible schooling options. Socio-economic Support: Implement programs that address the socioeconomic disparities influencing school selection, such as subsidies or financial aid for lower-income families. Community Engagement: Foster a collaborative approach involving parents in the educational policies and school management decision-making process.

This study acknowledges several limitations: Contextualization of the findings is specific to Visakhapatnam City and may not be generalizable to other urban areas with different socio-economic contexts. The cross-sectional nature of the study limits the ability to observe changes over time in contrast to longitudinal studies. Moreover, the snowball sampling method and self-reported data from parents may affect the results in potential biases.

In conclusion, this study highlights the critical need for a holistic approach to urban planning and school infrastructure development. Policymakers, urban planners, and educators must collaborate to create neighbourhood-level interventions that provide quality education and foster a sense of community. By addressing the demographic, socio-economic, and non-spatial attributes influencing school selection, we can ensure equitable access to education and contribute to the sustainable development of urban areas.

Acknowledgements

The author acknowledges the participants of the study for their valuable responses. Further, the author also acknowledges the municipal administration of Visakhapatnam City for facilitating the necessary support.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of Interest

The author(s) declare(s) no conflicts of interest.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author, Ram Sateesh Pasupuleti.

Institutional Review Board Statement

Not applicable.

CRediT author statement:

Conceptualization: R.R., R.S.P.; Data curation: R.R.; Formal analysis: R.R.; Funding acquisition: -; Investigation: R.R.; Methodology: R.R.; Project administration: R.S.P.; Writing—original draft: R.R.; Writing—review and editing: R.R., R.S.P. All authors have read and agreed to the published version of the manuscript.

References

Alsuiadi, F. A. (2015). Reasons influencing selection decision making of parental choice of school. International Journal of Research in Education and Science (IJRES), 2(1), 201–211. https://doi.org/10.21890/ijres.57410

Aronovici, C. (1939). PERRY, CLARENCE ARTHUR. Housing for the Machine Age. Pp. 261. New York: Russell Sage Foundation, 1939. $2.50. The ANNALS of the American Academy of Political and Social Science, 205(1), 159–160. https://doi.org/10.1177/000271623920500146

Beavis, A. (2004, January 01). Why Parents Choose Public or Private Schools. Research Developments, 12(12) https://research.acer.edu.au/resdev/vol12/iss12/3.

Bo Hu, Jun Saho, M. P. (2006). Pseudo-R 2 in Logistic Regression Model. Statistica Sinica, 16(3), 847–860.

Broberg, A., & Sarjala, S. (2015). School travel mode choice and the characteristics of the urban built environment: The case of Helsinki, Finland. Transport Policy, 37, 1–10. https://doi.org/10.1016/j.tranpol.2014.10.011

Burgess, S., Greaves, E., Vignoles, A., Wilson, D. (2014). What Parents Want: School Preferences and School Choice. The Economic Journal, 125(587), 1262–1289. https://doi.org/10.1111/ecoj.12153

Chica-Olmo, J., Rodríguez-López, C., & Chillón, P. (2018). Effect of distance from home to school and spatial dependence between homes on mode of commuting to school. Journal of Transport Geography, 72, 1–12. https://doi.org/10.1016/j.jtrangeo.2018.07.013

Chillón, P., Panter, J., Corder, K., Jones, A. P., & Van Sluijs, E. M. F. (2015). A longitudinal study of the distance that young people walk to school. Health and Place, 31, 133–137. https://doi.org/10.1016/j.healthplace.2014.10.013

Crane, R., & Crepeau, R. (1998). Does neighborhood design influence travel? A behavioral analysis of travel diary and GIS data. Transportation Research Part D: Transport and Environment, 3(4), 225–238. https://doi.org/10.1016/S1361-9209(98)00001-7

Easton, S., & Ferrari, E. (2015). Children’s travel to school—the interaction of individual, neighbourhood and school factors. Transport Policy, 44, 9–18. https://doi.org/10.1016/j.tranpol.2015.05.023

Faulkner, G. E. J., Buliung, R. N., Flora, P. K., & Fusco, C. (2009). Active school transport, physical activity levels and body weight of children and youth: A systematic review. Preventive Medicine, 48(1), 3–8. https://doi.org/10.1016/j.ypmed.2008.10.017

Glazer, N. (1959). The School as an Instrument in Planning. Journal of the American Institute of Planners, 25(4), 191–196. https://doi.org/10.1080/01944365908978333

Hastings, J., Kane, T., & Staiger, D. (2005). Parental Preferences And School Competition: Evidence From A Public School Choice Program. National Bureau of Economic Research. https://doi.org/10.3386/w11805

Iram Naila, Hussain, Z., Anwar, S., Hussain, I., & Waqar Akram. (2008). Determinants of Child School Choice in Punjab: Policy Implications. European Journal of Scientific Research, 23(2), 285–293.

Joshi, P. (2014). Parent decision-making when selecting schools: The case of Nepal. Prospects, 44(3), 411–428. https://doi.org/10.1007/s11125-014-9319-9

Kingdon, G. G. (2007). The progress of school education in India. Oxford Review of Economic Policy, 23(2), 168–195. https://doi.org/10.1093/icb/grm015

Klees, S. J. (2010). School choice international: exploring public-private partnerships, Globalisation, Societies and Education. Globalisation, Societies and Education, 8(1), 159–163. https://doi.org/10.1080/14767720903574140

Krishna, V. M., & Naidu, B. A. (2005). A comparative study of private and public secondary schools in Anantapur and Guntakal municipalities (Master's thesis). Sri Krishnadevaraya University.

Lawhon, L. L. (2014). Neighborhood Unit. Encyclopedia of Quality of Life and Well-Being Research, 4335–4337. https://doi.org/10.1007/978-94-007-0753-5_3335

Macdonald, L., Mccrorie, P., Nicholls, N., & Olsen, J. R. (2019). Active commute to school: does distance from school or walkability of the home neighbourhood matter? A national cross-sectional study of children aged 10-11 years, Scotland, UK. BMJ Open, 9(12), e03328. https://doi.org/10.1136/bmjopen-2019-033628

McDonald, N. C. (2007). Children’s mode choice for the school trip: The role of distance and school location in walking to school. Transportation, 35(1), 23–35. https://doi.org/10.1007/s11116-007-9135-7

Mchugh, M. L. (2013). The Chi-square test of independence. Biochemia Medica, 143-149. https://doi.org/10.11613/BM.2013.018

Mitra, R. (2013). Independent Mobility and Mode Choice for School Transportation: A Review and Framework for Future Research. Transport Reviews, 33(1), 21–43. https://doi.org/10.1080/01441647.2012.743490

Mousumi, M. A., & Kusakabe, T. (2022). Private school choice among Muslim parents: the public–private school decision in Delhi, India. London Review of Education, 20(1). https://doi.org/10.14324/LRE.20.1.25

Mumford, L. (1949). Planning for the Phases of Life. The Town Planning Review, 20(1), 5. https://doi.org/10.3828/tpr.20.1.q57u3u2w37h8816k

Ota, M., & Moffatt, P. G. (2006). The within-household schooling decision: a study of children in rural Andhra Pradesh. Journal of Population Economics, 20(1), 223–239. https://doi.org/10.1007/s00148-005-0033-z

Parruck Chanda, R., & Ghosh, A. (2014). Indian School Education System An Overview. The British Council of India, New Delhi.

Retrieved at https://issuu.com/britishcouncilindia/docs/indian_school_education_system_-_an_9d4e4afb7b85fd/2

Sabin, L. D., Behrentz, E., Winer, A. M., Jeong, S., Fitz, D. R., Pankratz, D. V., Colome, S. D., & Fruin, S. A. (2004). Characterizing the range of children’s air pollutant exposure during school bus commutes. Journal of Exposure Science & Environmental Epidemiology, 15(5), 377–387. https://doi.org/10.1038/sj.jea.7500414

Schneider, M., & Buckley, J. (2002). What Do Parents Want from Schools? Evidence from the Internet. Educational Evaluation and Policy Analysis, 24(2), 133–144. https://doi.org/10.3102/01623737024002133

Sidharthan, R., Bhat, C. R., Pendyala, R. M., & Goulias, K. G. (2011). Model for Children’s School Travel Mode Choice: Accounting for Effects of Spatial and Social Interaction. Transportation Research Record, 2213(1), 78-86. https://doi.org/10.3141/2213-11

Singh, A. (2015). Private school effects in urban and rural India: Panel estimates at primary and secondary school ages. Journal of Development Economics, 113, 16–32. https://doi.org/10.1016/j.jdeveco.2014.10.004

Singh, N., & Vasudevan, V. (2018). Understanding school trip mode choice – The case of Kanpur (India). Journal of Transport Geography, 66, 283–290. https://doi.org/10.1016/j.jtrangeo.2017.12.007

Singh, S. K. (2017). Road Traffic Accidents in India: Issues and Challenges. Transportation Research Procedia, 25, 4708–4719. https://doi.org/10.1016/j.trpro.2017.05.484

Srinivasan, M. (2010, August 24). How traffic congestion affects school students - The Hindu. The Hindu. https://www.thehindu.com/news/cities/chennai/How-traffic-congestion-affects-school-students/article16142542.ece

Sun, W., Guo, D., Li, Q., & Fang, H. (2021). School runs and urban traffic congestion: Evidence from China. Regional Science and Urban Economics, 86, 103606. https://doi.org/10.1016/j.regsciurbeco.2020.103606

Sun, Y., Liu, Y., & Tao, F. B. (2015). Associations between active commuting to school, body fat, and mental well-being: Population-based, cross-sectional study in China. Journal of Adolescent Health, 57(6), 679–685. https://doi.org/10.1016/j.jadohealth.2015.09.002

Tetali, S., Edwards, P., & Roberts, G. V. S. M. I. (2016). How do children travel to school in urban India? A cross-sectional study of 5,842 children in Hyderabad. BMC Public Health, 16(1), 1–7. https://doi.org/10.1186/s12889-016-3750-1

Tucker Shailey, S. G. (2012). 25% Reservation under the RTE: Unpacking the Rules in PAISA States.

Woodhead, M., Frost, M., & James, Z. (2013). Does growth in private schooling contribute to Education for All? Evidence from a longitudinal, two cohort study in Andhra Pradesh, India. International Journal of Educational Development, 33(1), 65–73. https://doi.org/10.1016/j.ijedudev.2012.02.005

Woessmann. L. (2000). Schooling Resources, Educational Institutions, and Student Performance: The International Evidence. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.234820

$D:\My Journal\Logo\kam logo.JPG$	Journal of Contemporary Urban Affairs
2024, Volume 8, Number 1, pages 232–248 Original scientific paper Parental Preferences and Non-Spatial Factors Influencing Home-to-school Distance in Urban India: Insights From Visakhapatnam ^1 Rohinikumar Reddy* , ²Assoc. Prof. Dr. Ram Sateesh Pasupuleti ¹^{& 2}Department of Architecture and Planning, Indian Institute of Technology Roorkee, India ¹E-mail: rrohinikumar@ar.iitr.ac.in , ²E-mail: ram.pasupuleti@ar.iitr.ac.in
ARTICLE INFO: Article History: Received: 25 April 2024 Revised: 18 June 2024 Accepted: 25 June 2024 Available online: 30 June 2024 Keywords: School Commute, Urban Education Planning, Home-to-School Distance, Parental School Choice, Multinomial Logistic Regression, Visakhapatnam.	ABSTRACT The school in the neighbourhood unit has been a significant and irreplaceable element in promoting active school commutes, and physical activity needs for children. However, urban areas in India are currently witnessing long school commutes. This study explores parental preferences for school selection in an urban Indian context, focusing on non-spatial attributes of schools associated with home-to-school distance. A questionnaire-based cross-sectional survey was conducted in Visakhapatnam, Andhra Pradesh, with a sample size of 409 families. Data were collected under four parameters: family demographics, socio-economics, neighbourhood environment, and non-spatial attributes of the schools. Pearson's chi-square (χ2) test of independence was adopted to identify the association between the dependent and predictor variables. Further, a multinomial logistic regression model was employed to predict the odds in home-to-school distance. These methods were chosen for their ability to provide robust and reliable results. The findings, which indicate that child age, number of school-going children, annual household income, school's board of affiliation, and education level significantly predict home-to-school distance, have practical implications. They suggest that policy interventions aimed at reducing school commute times should consider these non-spatial factors. Enhancing local school attributes could encourage shorter commutes, promoting healthier lifestyles for children. This understanding can guide the development of policies and interventions that effectively reduce school commute times.

This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution 4.0 International (CC BY 4.0) $C:\Users\Hourakhsh\Desktop\CC_By_2020_licnece1.jpg$ Publisher’s Note: Journal of Contemporary Urban Affairs stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
	JOURNAL OF CONTEMPORARY URBAN AFFAIRS (2024), 8(1), 232-248. https://doi.org/10.25034/ijcua.2024.v8n1-13 www.ijcua.com Copyright © 2024 by the author(s).
Highlights:		Contribution to the field statement:
- The school's non-spatial attributes are contextually specific. This study is the first attempt to investigate the role of non-spatial factors affecting home-to-school distance in India. - The non-spatial attributes of school environments significantly influence parents' preferences in school selection. - Statistical methods χ2 test of independence and MNL Regression analysis provide robust results in preferential choice studies.		This study expands theoretical perspectives on the neighbourhood school concept by incorporating non-spatial attributes of schools and socio-economic factors in parents' school selection preferences, emphasizing the need for integrated urban planning and community engagement to ensure equitable education access in developing urban areas.

1.2 Literature Study

1.2.1 Understanding the Indian School Education

1.2.2. Parents Preferences

3.1 Children studying within the neighbourhood vs outside