Revisiting the Use of the Proximity Lexicon in the Classroom for Immigrant Populations at Risk of Social Exclusion: Does It Really Help?

Abstract

This paper shows the convenience of using linguistic resources in foreign language teaching based on the proximity lexicon (dialectal and local) in order to facilitate the integration of socially vulnerable immigrant groups. The social barriers facing them intensify because of lack of knowledge of the linguistic varieties of the host countries and their functions. This study is a continuation of another one published in 2019 in which we presented a theoretical model of language teaching based on the gradual introduction of specific lexical material with the aim of facilitating sociocultural adaptation. The proximity lexicon associated to communal belonging meanings was considered a powerful social inclusion tool. We worked with two student samples to confirm the efficacy of the model. In a group, the dialectal strategy was employed in a programmed and conscious manner, whereas in the other one, the control group, the model was not followed. The results are based on the answers to questions in a wellbeing questionnaire designed ad hoc to evaluate the efficacy of the proposal. The group to which the proposed system was applied showed a greater degree of social wellbeing, satisfaction and adaptation.

1. Introduction

Population movements have been a constant throughout the history of mankind. Nevertheless, from the end of the XIX century, migratory flows have intensified and brought sociolinguistic, and unexpected, consequences which merit discussion and solution in case of emerging conflicts (Moreno-Fernández 2009; Fernández-Vítores 2013; Piller 2016; EUROSTAT 2017). In this paper, we present the early results of the application of a linguistic performance model, which aims, from the perspective of social dialectology, at facilitating the sociocultural integration of the more vulnerable immigrant groups by way of the teaching of the proximity lexicon (dialectal and local) in the Spanish as a foreign language (SFL) classroom. The epistemological foundation of this model was presented in Ávila-Muñoz (Forthcoming), although in that previous work, its practical efficacy was still to be demonstrated. The first section of this paper is dedicated to the synthesis of the more relevant theoretical fundamentals of the model and in the second one, the results of its application in the SFL classroom are submitted. This application took place in an NGO located in the city of Malaga working with migrants under risk of social exclusion. The results are based on the use and interpretation of wellbeing tests applied to a sample of students after their having completed their Spanish course. To insure the reliability of our proposal, we worked with two groups of informants to the first of which the proposed model was applied by way of showing the Proximity lexicon systematically and consciously, aiming at its transformation into a social integration tool. The second one—the control group—was not given language teaching based on the model.

2. The Linguistic Model of Social Integration Based on the Teaching of the Proximity Lexicon

2.1. Principles

As shown in Ávila-Muñoz (Forthcoming), the proposed model is based on the following principles:

• Principle of Group Belonging. Languages have units which function as solidarity links among members of speech communities because they connote sharing of cultural habits. These units are especially significative at the semantic-lexical level because of two well-documented motives in cognitive sociolinguistics (Moreno-Fernández 2013a). The first one being that each lexical variant carries associated underlying social meanings; in the case of localisms, this motive is linked to the history of the immediate sociocultural context. The second motive is that sociolinguistic conscience particularly sensible to these units, interprets them as symbols of social identity (Sheppard 2015; Regan et al. 2016; Dempster and Hargrave 2017; Stokes 2017).
• The Integration Principle. If these units are identified systematically and consciously, without prejudices, in foreign language teaching to immigrants, or even in general language classes to migrants who speak the language of the host community, we would be providing a sociocultural integration tool. This tool might also modify attitudes of negation or rejection by members of the host community (Ariolfo 2012; CIS 2015; Dempster and Hargrave 2017; Wheeler 2019).
• Principle of the usefulness of an entertaining multidimensional approach. The work with the specific vocabulary of the nearest social context assumes an additional motivation for both the teacher and the student. On the one hand, the teacher may know, respect and foster the use of this lexicon among his students. On the other hand, the characteristics of this lexicon lend support to the realisation of complementary workshops with the aim of backing the explanation of the connoted value of these linguistic units.

This approach can be applied to any foreign language teaching to migrants, but it acquires its whole dimension with those students who wish to remain and settle—either transiently or indefinitely—in a region of the host country. In addition to being an interesting approach to any situation involving the teaching of foreign languages or varieties of a language, we believe that these principles can lead to a pedagogical strategy of far-reaching sociocultural integration of vulnerable groups.

The proximity lexicon (dialectal and local) not registered in general dictionaries or marked as specific of concrete places, can be observed as complete forms (for example, in Andalusia, borrachuelo, ‘fried dough eaten during the Christmas/Easter seasons’) or as particular meanings of general forms (in Malaga, pitufo ‘small bread roll’). The proposal goes beyond the classroom and becomes a space for debate that would help to diminish the negative effects of the social marginalisation to which the more unprotected migrants are subjected (Moreno-Fernández 2013b; Sancho-Pascual 2013; Johnson and Berry 2014; Bravo-García et al. 2014). Our ultimate objective is therefore, the creation of a theoretical model of language teaching that on the bases of description, explanation and understanding of the sociolinguistic situation caused by these population flows, becomes an integration tool, above all for the groups more likely to suffer social exclusion. It is a question of facilitating the access to the knowledge of the language and the connoted usages associated to the employment of some elements typical of the linguistic variety nearest to the area where the learning process takes place.

One of the main advantages of this model is the creation of a discussion space among the agents involved in the language teaching process to migrant populations. From the practical problems facing teachers, we aim at making our proposal the beginning of a debate on the need to create teaching models based on tools whose function will be to show the linguistic variety to its full extent with the objective of facilitating social integration (Mallows 2014; Egli and Höchle 2016; Devereaux and Palmer 2019).

2.2. Theoretical Approaches and Operational Areas

The objective is the creation of a programming strategy which allows the incorporation of new resources to those normally used in language teaching. In order to achieve it, our model proposes procedures compatible with the elaboration of didactic programmings, which may even be in use. In sum, what is needed is the generation of a proposal with a verifiable efficacy easily put into practice by teachers, permitting both linguistic learning and the necessary social integration of the students.

Our proposal presupposes the convenience of using the Proximity lexicon in the language classroom, be it mother or foreign tongue, as a strategy for the social integration of migrants. However, this lexicon should be a reduced part of the whole lexicon; its verified social cohesion function, with the active connotations in the competence of the speakers of the local groups (Ávila-Muñoz 2017), makes its inclusion recommendable in the lexical selection models employed in the language teaching for vulnerable groups.

Our theoretical model with the inclusion of the Proximity lexicon in the language teaching classroom may be organised as a structure of concentrical crowns, built in five stages:

1.

Adequate selection of the lexicon. The common basis is formed by the general lexicon of very frequent use, not dependent on the communicative situation (Frequency Lexicon: Davies 2005; Lonsdale and Le Bras 2009) and availability, habitually associated to concrete situations (Lexical Availability Ávila-Muñoz and Villena 2010; Bombarelli 2005). From this common structure, our proposal assumes the introduction of some proximity elements, taking advantage of the linkage of contents of didactic units as presented in the system of concentrical crowns in Figure 1.

The always active nucleus present in the Frequency Lexicon is generally formed by grammatical words or those of scarcely precise semantic content: conjunctions, articles, prepositions and some nouns or verbs of polyvalent or ambiguous character (Ávila-Muñoz 1999). This nucleus is complemented as a function of the concrete communicative situation and of the words associated to it. Lexical availability refers to habitual vocabulary in concrete thematic contexts, for instance: family, human body, weather, and politics. The structure of the proximity lexicon can be observed as well as a wave system generated round a nucleus formed by this type of lexicon employed more often in the speech community. These waves distance themselves from the centre as the frequency of its terms diminishes as shown in Figure 2, which refers to the use of a series of localisms in the city of Malaga (Ávila-Muñoz 2017) whose gloss appears in

Table 1. Gloss of the localisms in Figure 2.

Localisms	Gloss
merdellón	‘not very fine, pretentions, vulgar’
piarda	‘play truant’ (UK), ‘play hookey / hooky’(USA)
guarrito	‘drill’
(d)amasquillo	‘apricot’
madrevieja	‘sewer’
aliquindoi	‘to be alert’
majarón	‘crazy, fool’
hueso cuqui	‘bone of the rump’
marengo	‘fish peddler’
naranja cachorreña	‘bitter orange’

.

From the model of general lexical representation suggested in Figure 1, the teacher should select the more appropriate words for each moment and social context in order to guarantee his students´ learning and make their sociocultural integration easier according to each type of education:

• On the one hand, in the case of the teaching of foreign languages, our proposal could be applied to those units related to all areas of communicative relevance proposed by the European Council in the European Common Frame of Reference for Languages (European Council 2002) or the oral linguistic recruitment guides presented by the American Council on the Teaching of Foreign Languages (ACTFL 2012).
• On the other hand, the Spanish non-university educational system could adapt, with certain ease, our proposal to the subject of Lengua Española (or the corresponding co-official one in those autonomous communities in which it exists) in order to work with students, both in the primary and secondary levels, especially in the latter.

Given the specificity of the proximity lexicon, an optimal selection of the communication domains or didactic units should be made, in which the proposal could be applied according to the most evident presence of this type of lexical items. For example, in the case of foreign language teaching, our experiment was done with the didactic unit related to the MCER proposal ‘Food and drink’. Together with general elements such as agua ‘water’, refresco ‘soft drink’, or limonada ‘lemonade’, we included others more specific to the area, which are shown in

Table 2. Proximity lexicon employed in the experiment (Ávila-Muñoz 2017).

Word	Gloss
agu<(j)>apala<(d)>a	‘sword fish’
arenca	‘herring’
vitoriano	‘small-sized anchovy’
borrachuelo	‘Christmas / Easter sponge cakes covered with rum´
<(d)>amasquillo	‘apricot’
pitufo	‘small piece of bread’
cachorreña	‘bitter orange’
sombra	‘white coffee’ coffee with some milk (+ coffee − milk)
nube	‘white coffee’ coffee with more mIlk´ (− coffee, + milk)

. In our case, these words come from a previous study in which as mentioned before, the proximity lexicon had a clear importance as a local identifier factor (Ávila-Muñoz 2017).

2.

The selection of the proximity lexicon should be adjusted to the model of concentrical crowns shown in Figure 2. The more frequent form—known and used by a great part of speakers of the local speech community—would be included in the centre and as they distance themselves from the initial nucleus, other forms of a lesser frequency of use would appear. For the design of the “Food and drink” didactic unit employed in our experiment, the model was fixed as shown in Figure 3.

The words included in the central circles or “crowns” of Figure 3, thus being more frequent and representative of the active Malaga proximity lexicon (nube, sombra, pitufo, borrachuelo, <d>amasquillo, vitoriano, arenca) should become the first lexical social integration tools and could be used for the intermediate levels proposed by ECFR (A2-B1) or the descriptors of ACTFL (Lower Intermediate, Intermediate). The more peripheral agujapala<(d)>a and cachorreña forms in Figure 3, are local meanings recognised by members of the Malaga speech community, but little or not used (passive lexicon), so that they could be assigned to higher levels although they were used in our experiment given the willing welcome by our students to the activity. In case the organisation of the thematic nuclei into expansion lexical waves should not be possible, the teacher could work with the more general model shown in Figure 2 and teach more widely used proximity elements without having to adjust obligatorily to a concrete didactic unit.

3.

After an adequate vocabulary selection, activities may be proposed concerning cultural information exchange about habits related to the themes used for the introduction of the lexicon. In our experiment, activities were organised in which the students’ gastronomic experiences and traditions were shared.

4.

These practical activities with the students should be backed by seminars, informal discussions or debates, and diverse activities outside the classrooms or teaching environments were encouraged. In our case, participation in gastronomic fairs or cooking workshops were organised, where the students were allowed to apply in a practical way what had been presented theoretically.

5.

Finally, it is necessary to share and put into practice the knowledge thus acquired so as to evaluate the real efficacy of the Proximity lexicon as a tool towards sociocultural integration.

3. Hypothesis and Design of Variables

With the object of verifying the efficacy of our proposal, during 2018 we contacted several NGOs working with migrant groups under risk of social exclusion in the city of Malaga. We came to the conclusion that Málaga Acoge (Malaga Welcomes You) was the one that fulfilled all the requirements to implement our model:

• It had active courses of Spanish as a foreign language.
• The students were classified into groups according to their level of knowledge of the Spanish language. The criteria and proposals of ECFR were employed in the evaluation of their idiomatic command of the language.
• Every level had a sufficient number of students.
• Additionally, there was more than one group for each different level of knowledge of the language.

In a first approximation, we believed it convenient to work with Spanish B1 level students as the course had already started at the beginning of our research. We did not have more than a quarter of them to carry out our experiment and it seemed convenient to put our model into practice at that intermediate level. It was also important to keep a control group which during their learning process, did not receive any stimulus by way of the Proximity lexicon and if it did, the stimulus was not to be given in the systematic and controlled manner required by our research.

Both groups constitute the independent variable of our work and from its design, we were able to build the main hypotheses of the present research:

• Hypothesis of personal satisfaction. After the implementation of the didactic model which supports this work, there will be a statistically superior and significative difference in the experimental group in contrast with the control group in its perception of local integration, wellbeing and personal satisfaction.
• Hypothesis of integration. Likewise, it is to be expected that the positive perception of vital wellbeing among the individuals which form the experimental group may lead them to formulate plans of possible plans of a future and integration in the host community to a greater degree than the members of the control group.

With the objective of confirming both hypotheses, we carried out an analysis of correlations between dependent and independent variables with the results which are presented in the following chapter. The dependent variables were built on the bases of the answers given by the informants to the items included in the sociological questionnaire of wellbeing (Appendix A). The results were correlated with the independent variable (group to which the individual belonged: experimental group, Group A; control group, Group B). The idea was to establish a cause-effect relation between the variables under study so that the belonging to each group should justify the level of satisfaction as formulated initially in the hypotheses.

We had a total sample of 40 students. Half of them (N = 20) were offered the learning system which contained the “Food and drink” unit. We chose this unit because of its traditional use in foreign language teaching. In addition, a previous study gave us enough local terms connected with this unit. As we will mention later, it is important a previous study of specific situations to use the proposal model in other didactic units. At the same time, as mentioned, a control group of 20 students was kept to which the model was not applied (Group B), even though the rest of educational process was identical (same learning system, same didactic unit Food and drink). The main difference was the systematic way that the experimental group learnt local lexicon. As obvious, the teachers in the control group could use this lexical forms eventually. The difference was the systematic and conscious way that defines the experimental group.

Throughout the last quarter of the course (from March to June 2018), the teachers under contract to the NGO, were asked to apply the proximity-based model to the members of Group A. Naturally, the teachers were previously informed of the nature of the activity and the lexicon model methodology better suited to be implemented with Group A. Furthermore, a group of students who had been previously informed and trained, volunteered as auxiliary instructors during that quarter. The auxiliary group helped to organise workshops and activities related to the selected didactic unit: local cooking workshops, excursions to gastronomic provincial fairs and an end-of-course celebration, where local food and drinks were exchanged. The exchange also included food and drinks from the migrants’ homelands. As mentioned before, the lexical items employed came from the previous study and selection (Ávila-Muñoz 2017) and were shown in Table 2.

At the end of the course, the students were asked to fill the wellbeing questionnaire shown as Appendix A. In the following section comments are given on the more significative results from the analysis of those questionnaires, that were created ad hoc to improve the actual research from previous studies (Diener et al. 1985; Atienza et al. 2000; Pons et al. 2002)

4. Sample

All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of ECOPASOS (FFI2015-68171-C5-1-P).

The main social and personal characteristics of our informants are synthesised in

Table 3. Characteristics of the sample. Gender and origin.

		Gender		Origin			Total
		Male	Female	Europe	Africa	Asia
	Group A	12	8	1	16	3	20
	Group B	12	8	0	18	2	20
Total		24	16	1	34	5	40

,

Table 4. Characteristics of the sample. Completed studies.

		Studies That Were Completed					Total
		No School Attendance Whatsoever	No, but (s)he Attended School	Primary	Secondary	Professional
	Group A	4	1	14	1	0	20
	Group B	8	4	7	0	1	20
Total		12	5	21	1	1	40

and

Table 5. Recodified educational level. Distribution by groups.

		Recodified Education		Total
		Level 0	Level 1
	Group A	5	15	20
	Group B	12	8	20
Total		17	23	40

and provide the description of our sample.

Most of the migrants came from Africa, although in both groups there were people from Asia and only one was from a European country in Group A (Table 3). The number of males was somewhat higher than that of females, but the imbalance was identical in both groups (M = 12, F = 8).

A recodified version more adjusted to the informants’ educational reality was done based on the data in Table 4. Thus, the values were reorganised into two variants according to the degree of completion of some educational stages (Level 1, N = 23) or not (Level 0, N = 17). We believe that the reorganisation shown in Table 5 is a better representation of our informants’ situation than that of Table 4 and it is more adjusted to the prevalent academic situation in the western world.

Most of the individuals in Group A fit in Level 1 (N = 15), the opposite situation occurring in Group B, since almost all its members belong to educational level 0.

Table 6. Educational level according to gender.

Gender		Recodified Education		Total
		Level 0	Level 1
	M	6	18	24
	F	11	5	16
Total		17	23	40

shows that most males indicated that their study rank corresponded to Level 1 in our recodification (N = 18), while most females belonged to Level 0 (N = 11).

Most of the members of the sample had arrived to Malaga between a year and less than a year (Group A = 17; Group B = 15) and only one had been more than five years in the city. The rest had been between 1 and 5 years in Malaga (Group A = 3; Group B = 4). Almost all had had very frequent contact with their country (between two or more times every week and every day). Only 2 informants from Group A indicated that they kept very sporadic contact with their country (more than twice monthly). Half of the sample did not answer the questions related to the work they had (Group A = 11; Group B = 9), which made us think they were unemployed or in an illegal situation at the time of the interview. It is relevant to point out that of those who answered those questions, almost all expressed that they were very little/little satisfied with their work and only one person of Group B stated that (s)he was very satisfied with her or his employment situation. Additionally, it was lamentable that 30 of our informants had suffered some racist or xenophobic incidents since the time they arrived in Spain (Group A = 16; Group B = 14), whereas 4 classified such incidents as grave and 26 as Not grave.

Most of the informants indicated that they shared their free time only with people of the same origin or culture (Group A = 13; Group B = 17). However, 6 people from Group A and only 1 from Group B also socialised with Spaniards. Almost none of the migrants belonged to associations or social groups (Group A = 18; Group B = 18) and only 4 participated in associations, which were anyway formed by people of their own origin or culture (Group A = 2; Group B = 2).

As can be appreciated by the data so far, the initial conditions of the members of Groups A and B were very similar, except for their educational level. Nonetheless, their education showed no correlation or a very weak one with the variables we will use further on to measure the wellbeing shown by the migrant towards the population and their host region or area (see below, regression analysis by steps).

5. From Theory to Practice. Does the Model Work?

It is important for the objectives of our work that 9 people from Group A indicated that since they began their study of Spanish, they had fewer integration problems in contrast with only 3 informants who chose the same answer in Group B. This part of our data may turn out to be very significative because until the implementation of the proximity model in March 2018, the learning conditions of both groups were the same: identical language learning procedures in so far as contents, teaching and methods (at any rate, without the systematic employment of the Proximity lexicon as a working tool). In Group B, only 2 people stated they had no integration problems in contrast with 7 who chose the same option in Group A. Likewise, we believe that data in

Table 7. Integration difficulties with Malaganians *.

		Integration Difficulties with Malaganians					Total
		0	1	2	3	4
	Group A	7	9	0	3	1	20
	Group B	2	3	8	6	1	20
Total		9	12	8	9	2	40

0: I have no problems. 1: Since I study Spanish, it is easier for me to integrate. 2: No. I do not look for this type of contacts. 3: No. I just prefer to be with people who share my culture. 4: Yes, I have many problems to integrate. * Data obtained from the answers to question number 25 in the questionnaire.

are relevant because we could interpret that the teaching method based on the Proximity lexicon as we propose it, could have some incidence on the social integration ease of the immigrants in the local communities.

Notice how 8 members of Group B did not even look for contacts; no member of Group A chose this option. Besides, the number of people from Group B (N = 6) doubles that of those in Group A (N = 3) who prefer to be with members of the same culture. In Group A, we find 16 people who state they have no integration problems or that they have been reduced since they study Spanish (options 0 + 1, N = 16) in contrast with only 5 in Group B.

When we analyse the answers referring to students´ integration with Malaga values and customs, the answers pointed in the same direction as shown in

Table 8. Integration in Malaga society *.

		Degree of Integration in Malaga					Total
		None	Very Little	Little	Enough	Quite a Lot
	Group A	0	3	11	5	1	20
	Group B	3	6	10	1	0	20
Total		3	9	21	6	1	40

* Data obtained from the answers to question number 31 in the questionnaire.

.

By adding the answers to None and Very little we observe that 9 people belong to Group B and only 3 to Group A (besides, in this group, nobody chose the option None). Additionally, 6 members of Group A indicated that they were between Enough and Quite a lot integrated in the Malaga society and only 1 person from Group B agreed with that position.

With regard to the migrants´ intentions in the short and the medium term, the data again consolidate the tendencies indicated: the members of Group A are the ones who in the majority consider, as a probable option, to remain in Malaga both in the short term (less than 5 years: Group A = 6; Group B = 1) as well as in the medium one (more than 5 years: Group A = 5; Group B = 0). Nevertheless, the truth is that most of the migrants we have studied, either do not know what they will do in the short/medium term or have in mind to go to some other countries.

We have left for the end of our analysis some of the items which in our opinion, will be the most useful in the evaluation of the efficacy of the proposed teaching model based on the proximity lexicon as a social integrational tool. We are referring to the gradation migrants make of (A) their feelings towards the assertions shown in items 39–41 of our questionnaire (

Table 9. Effect of the learning group in the subjective evaluation of life in Malaga *.

		In the Majority of Things, My Life in Malaga Is Positive				Total
		0	1	2	3
	Group A	2	5	8	5	20
	Group B	7	9	4	0	20
Total		9	14	12	5	40

* Data from the answers to question number 39 in the questionnaire.

,

Table 10. Effect of the learning group on the subjective evaluation of life in Malaga.

		Positive Evaluation of Life in Malaga
Group A	Mean	2.8000
	Typ. Dev.	0.95145
Group B	Mean	1.8500
	Typ. Dev.	0.74516
Total Mean		2.3250
Typ. dev.		0.85342
Sig. 0.001

and

Table 11. Effect of the learning group in the subjective evaluation of life conditions in Malaga *.

		My life Conditions in Malaga Are Adequate				Total
		0	1	2	3
	Group A	2	4	12	2	20
	Group B	16	4	0	0	20
Total		18	8	12	2	40

* Data obtained from answers to question number 40 in the questionnaire.

) and (B) their level of wellbeing in Malaga (

Table 12. Effect on the learning group on the subjective evaluation of life conditions in Malaga.

		Subjective Evaluation of Life Conditions in Malaga
Group A	Mean	2.7000
	Typ. Dev.	0.80131
Group B	Mean	1.2000
	Typ. Dev.	0.41039
Total	Mean	1.9500
	Typ. Dev.	0.67436
Sig. 0.000

, item 42 of the questionnaire). To evaluate these 4 items, the informants had to place them in a continuous scale from less to more (0–6).

The only grade 3 evaluations shown in Table 9 belong to Group A subjects (N = 5); notice that 16 people from Group B chose the lower levels of the scale (grades 0 and 1) in contrast to 7 of Group A. Additionally, only 4 people from Group B chose option 2, which was half of those in Group A (N = 8).

The media analysis shows a weak signification index, although with an Eta² of low intensity (Table 10).

The Mann–Whitney test was applied for differences of means in independent samples

Measures of Association	Eta	Eta2
Group * Positive Evaluation of life in Malaga	0.506	0.256

Similar tendencies are observed when analysing the data from the evaluation of life conditions in Malaga (Table 11). Nevertheless, on this occasion, the values of the means and the measures of association are very significative (Table 12). It calls our attention that all members of Group B chose the lowest proposed grades (N = 20, grades 0 and 1). Only 6 people from Group A opted for the same levels, in contrast with the 14, who chose 2 or 3.

The Mann–Whitney test was applied for mean differences in independent samples

Measures of Association	Eta	Eta2
Group * Sentence2	0.789	0.622

To think of moving dear relatives to a new country to start a new life is a very difficult decision, which requires a high degree of certainty and determination. However, as can be observed (

Table 13. Effect of the learning group on expectations of moving the family to Malaga.

		If It Depended on me, I Would Bring My Family to Malaga				Total
		0	1	2	3
	Group A	0	1	18	1	20
	Group B	16	4	0	0	20
Total		16	5	18	1	40

* Data obtained from the answers to question number 41 in the questionnaire.

) most of the subjects in Group A chose the highest option in this aspect (grade 2, N = 18), while the 20 subjects of Group B chose option 0 (N = 16) or 1 (N = 4). Nobody in this group chose option 3 in contrast with 1 person in Group A.

The strong signification values obtained from the media analysis led to a possible positive effect of the proposed teaching model (

Table 14. Expectations of moving the family to Malaga.

		Expectations of Moving the Family to Malaga
Group A	Mean	3.0000
	Typ. Dev.	0.32444
Group B	Mean	1.2000
	Typ. Dev.	0.41039
Total	Mean	2.1000
	Typ. Dev.	0.38845
Sig. 0.000

).

The Mann–Whitney test was applied for the media differences in independent samples

Measures of association	Eta	Eta2
Group * Expectations of moving the family to Malaga	0.959	0.920

Table 15. Effect of the learning group on the general level of wellbeing in Malaga *.

		General Level of Wellbeing in Malaga					Total
		0	1	2	3	4
	Group A	2	4	5	6	3	20
	Group B	4	11	4	1	0	20
Total		6	15	9	7	3	40

* Data obtained from the answers to question number 42 of the questionnaire.

shows very similar results, although in this occasion, the tendency towards a more positive evaluation of wellbeing in Malaga seems even more evident among members of Group A. Nobody in Group B indicated grade 4 and only one chose option 3. This option was selected by 6 members of Group A, who also indicated their preference for option 4 three times. More than half of the subjects of Group B opted for the lower grades of the scale (0 or 1, N = 15) while only 6 did the same in Group A.

As

Table 16. Effect of the learning group on the sensation of general wellbeing.

		General Wellbeing
Group A	Mean	3.2000
	Typ. Dev.	1.23969
Group B	Mean	2.1000
	Typ. Dev.	0.78807
Total	Media	2.6500
	Typ. Dev.	0.98976
Sig. 0.026

shows, the signification and association measures were more moderate in this case.

The Mann–Whitney test was applied for media differences in independent samples

Measures of association	Eta	Eta2
Group * Wellbeing in Malaga	0.515	0.265

The signification index of the positive correlations in

Table 17. Learning group and wellbeing indicators. Spearman correlations.

	Sentence 1	Sentence 2	Sentence 3	Wellbeing in Malaga
Group	0.491(**)	0.778(**)	0.934(**)	0.468(**)
	Sig. 0.001	Sig. 0.000	Sig. 0.000	Sig. 0.002
N = 40 ** The correlation is significative at the level of 0.01
Sentence 1: In most things, my life in Malaga is positive Sentence 2: The conditions of my life are adequate Sentence 3: If it depended on me, I would bring my family to live in Malaga

is high enough in all cases so that the belonging to learning groups can be considered a relevant explanatory factor of the variables that measure the migrants´ degree of wellbeing in their host community. The form of the variables correlated in Table 17 were recodified on the basis of a fictitious variable in which Group B = 0 and Group A = 1. Variables Sentence 1, Sentence 2, Sentence 3 and Wellbeing in Malaga were measured on a scale of gradation from 0 (completely in disagreement) to 6 (completely in agreement).

At the end, we made a recodification of some of the items of the questionnaire, whose scales were compatible and from them, two new variables were made in order to measure: (1) the informant´s global subjective satisfaction (Individual Subjective Satisfaction, ISS) and (2) the individual sensation of general wellbeing (General Wellbeing Index, GWI).

The first case deals with the original questionnaire variables, which measure on a scale from minor to major (1 to 5) the integration and opening of the analysed migrants to the host region and its inhabitants. Variable 25 was recodified to adapt it to the new scale.

23. You have meetings in your free time

• Only with members of my family
• Only with people of my origin or culture
• Never with people of my origin or culture, except with Spaniards
• With every type of people, including Spaniards

24. You participate in associations or groups in free time

• No
• Yes, only with people of my origin or culture
• Yes, but never with people of my origin or culture
• Yes, with every type of people, except with Spaniards
• Yes, with every type of people, including Spaniards

25. Do you have difficulties to integrate with Malaganians or groups of them? Think, for example. if you feel comfortable when you are with these people and consider you an equal or if, on the contrary, when you are with these people you feel a foreigner

• Yes, I have many difficulties to integrate
• No, but I only prefer to be with people of my zone or culture
• No, I do not look for this type of contact
• Since I study Spanish I find it easier to integrate
• I do not have any problems

28. How do you evaluate the native population of Malaga?

• Very negatively
• Negatively
• I am indifferent towards them
• Positively
• Very positively

31. In what way do you feel integrated with the customs of the Malaga society? (traditions like Easter, Carnival, etc.)

• Not at all
• Very little
• Little
• Enough
• Quite a lot

33. What are your intentions now?

• To make money and go back in less than a year
• To make money to build a better life in my country of origin
• To start a new life in another country
• To go back and stay with my family
• To stay in Malaga

34. What do you think you will do in the next 5 years?

• Going back to my country
• Moving to another country
• Moving to another Spanish autonomous community
• Moving to another city in Andalusia
• Staying in Malaga

35. What do you think of doing after the next 5 years? (As a life project)

• Going back to my country
• Moving to another country
• Moving to another Spanish autonomous community
• Moving to another city of Andalusia
• Staying in Malaga

The resulting variable of the reformulation of the previous ones (ISS calculated as a function of the number of times informants opted for the values of the scale 1–5 in all the considered variables) showed a normal distribution with some differences of means of a low potency (

Table 18. Effect of the learning group in the Individual Subjective Sensation (ISS).

		ISS	Punctuation Z
Group A	Mean	3.9500	0.5071618
	Typ. Dev.	1.46808	0.91260825
Group B	Mean	2.0500	−0.3631248
	Typ. Dev.	1.05006	0.65275386
Total	Mean	3.2500	0.0720185
	Typ. Dev.	1.44559	0.89862964
Sig. 0.001

).

The Mann–Whitney test was applied for the differences of means in independent samples

Measures of association	Eta	Eta2
ISS * Group	0.490	0.240
Z punctuation (ISS) * Group	0.490	0.240

The second recodification was done from the individual original variables, which measured on a 0–6 scale certain wellbeing parameters related to the informants´ stay in Malaga (variables 39–42 in the questionnaire). In this occasion, the new recodified scale in three levels of intervals according to the informants´ original answers (0–2 = Level 0; 3 = Level 1; 4–6 = Level 2). The differences of means show results of very high significance, now with values de Eta² of high potency (

Table 19. Effect of the learning group in the General Wellbeing (GWI).

GWI	Mean	Typ. Dev.
Group A	1.9500	0.22361
Group B	1.0000	0.00000
Total	1.4750	0.50574

).

The Mann–Whitney test was applied for differences of means in independent samples

Measures of Association
	Eta	Eta2
GWI * Group	0.872	0.761

Test of independent samples

GWI		Levene Test for Equality of Variances		T test for the Equality of Variances
		F	Sig.	t	df	Bilateral Sig.	Differences of Means
	Equal variances have been assumed	4.457	0.041	−19.000	38	0.000	−0.95000
	Equal variances have not been assumed			−19.000	19.000	0.000	−0.95000

Our significative results, and a final analysis of regression by steps, helped us to design a statistical model, which confirms: (1) the high correlation observed between group belonging and the variable GWI and (2) the weak correlation between the variables gender and education and the dependent variable (GWI, both variables were excluded from the model,

Table 20. Regression analysis.

Introduced/Eliminated Variables
Model	Introduced Variables	Eliminated Variables	Method
1	Group	Gender Education	By steps (criterion: Prob. of F to enter <= 0.050. Prob. of F to leave >= 0.100).

Dependent Variable: IBG.

). Thus, we discarded that the internal group imbalance with regard to the educational level of their members (Table 5 and Table 6) was camouflaging a possible covert effect of this variable on the results.

Eliminated Variables
Model		Beta in	t	Sig.	Partial Correlation	Colinearity Statistics
						Tolerance	VIF	Minimal Tolerance
1	Gender	0.041 b	0.813	0.422	0.132	1.000	1.000	1.000
	Recodified Education	−0.029 b	−0.536	0.595	−0.088	0.875	1.143	0.875

Dependent Variable: GWI. ^b Predictors in the model: (Constant), Group.

Eliminated Variables
Model		Beta in	t	Sig.	Partial Correlation	Colinearity Statistics
						Tolerance	VIF	Minimal Tolerance
1	Gender	0.041 b	0.813	0.422	0.132	1.000	1.000	1.000
	Recodified Education	−0.029 b	−0.536	0.595	−0.088	0.875	1.143	0.875

Dependent Variable: GWI. ^b Predictors in the model: (Constant), Group.

Summary of the Model
Model	R	R square	Corrected R square	Standard error in the estimate	Durbin-Watson
1	0.951 a	0.905	0.902	0.15811	2.108

^a Predictors: (Constant), Group. Dependent Variable: GWI.

Summary of the Model
Model	R	R square	Corrected R square	Standard error in the estimate	Durbin-Watson
1	0.951 a	0.905	0.902	0.15811	2.108

^a Predictors: (Constant), Group. Dependent Variable: GWI.

ANOVA a
Model		Sum of Squares	df	Mean Square	F	Sig.
1	Regression	9.025	1	9.025	361,000	0.000 b
	Residual	0.950	38	0.025
	Total	9.975	39

^a Dependent Variable: GWI, ^b Predictors: (Constant), Group.

ANOVA a
Model		Sum of Squares	df	Mean Square	F	Sig.
1	Regression	9.025	1	9.025	361,000	0.000 b
	Residual	0.950	38	0.025
	Total	9.975	39

^a Dependent Variable: GWI, ^b Predictors: (Constant), Group.

Coefficients
Model		Non-Standardised Coefficients		Typified Coefficients	T	Sig.
		B	Standard Error	Beta
1	(Constant)	1.000	0.035		28.284	0.000
	Group	0.950	0.050	0.951	19.000	0.000

Dependent Variable: GWI.

Coefficients
Model		Non-Standardised Coefficients		Typified Coefficients	T	Sig.
		B	Standard Error	Beta
1	(Constant)	1.000	0.035		28.284	0.000
	Group	0.950	0.050	0.951	19.000	0.000

Dependent Variable: GWI.

6. Discussion and Final Reflections

One of the major obstacles migrants have when they arrive at their destinations is their lack of knowledge of the local language and customs of the countries chosen for settlement. Nonetheless, in this work we have aimed at transforming this difficulty into the possibility of providing a better adaptation by way of a linguistic learning model backed by the following fundamentals, a consequence of what we have exposed so far:

• There is an objective need to formulate proposals leading to the successful sociocultural integration of immigrants (particularly the more unprotected and vulnerable among them) from the language classrooms.
• The challenge consists of supplying the lack of linguistic knowledge with inclusive programmes which, with respect to the nearest traditions, manage to promote the immigrants´ sociocultural integration in a multidimensional way and their acceptance by the host communities
• Consequently, it is a question of promoting educational programmes based on respect for the rights of the more vulnerable groups, which also deepens mutual understanding and acceptance of cultures and values of both the host communities and of the immigrant groups.
• Language has proved to be a powerful social connection between groups enjoying equality and that fact has been fully recognised by the main international institutions, which observe and study population and migratory flows (OECD/European Union 2015; Helke and Laczko 2015; Jahan 2016). Language teaching models should be adapted to include effective elements that facilitate this integration.
• Social dialectology is aware of a dynamic convergence process which is generalised among varieties of the same language (Auer 2005; Auer et al. 2005; Hinskens 1996; Kerswill 2003; Trudgill 2002; Villena 2005, among many others). This process strengthens the idea—widely accepted in the foreign language teaching field—of employing the standard variety as the basis of learning. However, the divergence processes play an important role not to be underestimated (Braunmüller et al. 2014; Villena and Vida 2017): the divergent variants reinforce the social solidarity links and can be used as integration tools when working with vulnerable groups (Ávila-Muñoz Forthcoming).
• Our proposal is compatible with the usual native and foreign language teaching processes. While complementing already elaborated didactic units, it could be also employed in the preparation of new ones. As the proximity vocabulary is restricted, the selection of the more adequate ones for each didactic unit will depend on each specific situation. The teachers are in charge of selecting and adding the relevant dialectal lexical elements to be introduced in the classroom gradually and adequately. The selection would have to be preceded by a study of the particular characteristics of each lexical units of the area for which glossaries and specialised dictionaries should be consulted. Furthermore, this study could also come from the knowledge teachers have of the sociolinguistic reality of the area in which they work. It is a question of trying to adjust the lexical selection to the model of concentrical crowns in each case as suggested in our proposal:
  • Inclusion of the proximity lexicon associated to specific thematic units or communication spaces when sufficient material exists. In these cases, the proximity vocabulary is to be presented and structured according to the model of concentrical crowns model, with a nucleus formed by extensively used and recognised elements in the local community—and thus, with a strong connoted value of social integration and group belonging conscience—it would be used at the most elementary: A2-B1 or Lower Intermediate-Intermediate for foreign language teaching; primary education in the case of the general teaching regime. In the higher levels, less-used vocabulary could be introduced in the speech community, although with the same social cohesion function.
  • Should the number of lexical elements be insufficient to execute the proposal as structure in concrete didactic units, the teacher should appeal to the general model of lexical selection (frequent lexicon, lexical availability, proximity lexicon) and include dialectal elements from the peripheral crown containing units with the required social integration function.
  • Lastly, the use of the proximity lexicon in the language classroom facilitates the realisation of complementary practical activities (workshops, informal meetings, debates), which may prove attractive to the students.

The results which have been shown in this paper after the implementation of the proposed model with a group of migrants as students of SFL, give room great hope. If we compare the results from the wellbeing questionnaire as answered by students after the end of the course according to our model with the answers of the members of the control group—who as such were not asked to do that course—we can assert that the model seems to work. Those who enjoyed the course with our model, showed a significantly higher degree of social integration, personal wellbeing and subjective satisfaction than those who followed a different methodology. The former are individuals with a more positive perception of their life conditions in their host community than the members of the control group, to the point that they even plan to bring their relatives to their actual location. This very positive attitude envisages commencing a new life in Malaga together with their family, which definitely indicates a considerably greater index of general well-being.

Summing up we can assert that we have created a foreign language teaching programme for migrant populations from the basis of a deep knowledge of the difficulties for the social integration of these groups. This programme justifies the use of the proximity lexicon in the language classroom as a sociocultural integration tool for migrant populations, but above all, for the more unprotected groups, who may find the employment of the proximity lexicon an entertaining and flexible manner to attain inclusion in the community. Although it is true that teachers may have appealed to these strategies previously in a more or less sporadic and unsystematic manner, from this point onwards, a collective space for discussion and work has been created, which justifies, structures, endorses and recommends this practice.