In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
707
De e minan s o O - a m Ac i i y Pa icipa ion
among Co on Fa me s in Punjab, Pakis an
Muhammad Amjed Iqbal
College o Economics and Managemen , Huazhong Ag icul u al Uni e si y,
Wuhan, China
Qing Ping
1
College o Economics and Managemen , Huazhong Ag icul u al
Uni e si y, Wuhan, China
Uma Ijaz Ahmed
College o Economics and Managemen , Huazhong Ag icul u al
Uni e si y, Wuhan, China
Adnan Nazi
College o Economics and Managemen , Huazhong Ag icul u al
Uni e si y, Wuhan, China
Abs ac
The s udy was conduc ed o de e mine he aspec a ec ing he a me in
aking pa in o - a m ac i i ies in h ee co on g owing dis ic s o Punjab
p o ince, Pakis an. Since o - a m ac i i ies ha e become an impe a i e pa o
income s a egies among u al amilies in de eloping coun ies like Pakis an.
The da a was documen ed om a o al o 180 co on a me s using mul is age
clus e sampling echnique. A bina y logis ic model was used o e alua e he
de e minan s mo i a ing he a me s o pa icipa e in di e en o - a m
ac i i ies. Va ious socio-economic ac o s we e ound signi ican ly associa ed
wi h p obabili y o imme sion in non- a m ac i i ies. The esul s o he model
e eal di e en ac o s like o al a ming a ea and a me s ha ing access o oad
we e signi ican o se e al business ac i i ies h ough odds a io 1.051 and
0.088 espec i ely. Though mo e expe ienced a me s wi h odds a io (1.063)
had mo e likelihood o labou ac i i ies. Las ly mo e educa ed a me s and la ge
amily size ha e highe p obabili y o go o se ices ype o o - a m ac i i ies
and hei odds a io es ima ed is 1.297, 2.069. These indings ha e essen ial
implica ions o policy, economic g ow h and de elopmen .
1
Co esponding au ho ’s email: qin[email p o ec ed]zau.edu.cn
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
708
Keywo ds: Ag icul u al households, income, de e minan s, o - a m
ac i i ies, pa icipa ion
Ci e his a icle: Iqbal, M. A., & Ping, Q. (2015). De e minan s o O - a m Ac i i y
Pa icipa ion among Co on Fa me s in Punjab, Pakis an. In e na ional Jou nal o Managemen ,
Accoun ing and Economics, 2(7), 707-718.
In oduc ion
In he pas he opinion was ha a ming households we e jus dependen on ag icul u e,
and didn’ need any o - a m ac i i ies. So he policy make s jus ocused on a ming
sec o . Bu du ing he las se e al yea s he e ha e been amassed indica ions which show
ha small a ming households do no ely on ag icul u e only. His o ically a me s ha e
ied o main ain hei ange o income ac i i ies in which o - a m ac i i ies ha e played
a i al ole (Ba e , Rea don, and Webb, 2001). Though Haggblade, Hazell, and Rea don
(2010) poin ed ou ha income om o he han ag icul u e sou ces has a sha e o abou
35 o 50 pe cen o he o al income o u al households in de eloping coun ies. The e
is an an icipa ion ha po ion o o - a m income will inc ease in he upcoming yea s,
mainly in he coun ies whe e he e is apid inc ease in popula ion and ha e inadequa e
ag icul u al sou ces a e a big h ea en o ag icul u e sec o (Haggblade, Hazell, and
Rea don, 2007). Almos he end o ecei ing income om non-ag icul u e sou ces is
inc easing in all coun ies i.e. Jolli e, (2004) indica ed ha in u al Ghana in 2004 nea
abou 74% o a ming households we e in ol ed in o - a m sec o , while ano he s udy
by (Fe nandez-Co nejo, 2007) na a ed ha 65% a m household we e a ached wi h o -
a m income sou ces and in Taiwan his a e has inc eased up o 75% espec i ely.
Albei he main easons o a me s’ in ol emen in o - a m sec o a e g ea e e u ns
and low isk o in es men in non-ag icul u e sec o (Kilic, Ca le o, Miluka, and
Sa as ano 2009). As De Jan y, Sadoule , and Zhu (2005) di ulged ha non- a m
employmen has a a o able e ec on ag icul u al p oduc ion, because he income
ob ained h ough his sec o can be used o in es ing on a m and will bene i he a me
o p ac ice a ming on ime. Mo eo e S ampini & Da is, (2009) o e old ha non- a m
employmen has in luenced he use o a iable inpu s in u al Vie nam, whe e a me s
ha ing o - a m employmen sou ces in es ed mo e on seeds, e ilize , ag icul u al
se ices and hi ed labou .
S ill o - a m ac i i ies ha e become a key agmen o a me ’s o e all income in he
de eloping coun ies. As ag icul u e sec o is qui e isky, because a m p oduc ion
dependen on many ac o s which a e un-con ollable by he a me . In spi e o his he e
is no gua an ee o bene icial e u ns om a ming. Many a me s ha e ied o adap
di e en sou ces o ea nings o ensu e hei a m income (De Jan y & Sadoule , 2001;
Ruben & Van den Be g, 2001; Haggblade e al., 2007). The e a e many easons o his
obse ed changing phenomenon which may be educ ion in a m income and willingness
o p o ec a m p oduc i i y agains di e en isks (Rea don, 1997; Ellis, 1998).
Doing o - a m wo k o gene a ing income is a way o sel -insu ing s a egy o he
a me s o s eng hen he households o e all income (Alasia, Wee sink, Bollman, and
C an ield, 2009). I is e y help ul o he a me in a condi ion when he has o ace loss
in ag icul u e due o any ac o which may be na u al o e c. Va ia ion among income
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
709
sou ces is one o he isk managemen s a egies o co e he isks. Rea don, Delgado,
and Ma lon (1992) indica ed ha household mus gi e a en ion o income di e si ica ion
as a s a egy o minimize he a ming income isks. I can play a posi i e ole in s abilizing
income and educe income inequali y among u al households. A a ie y o s udies ela ed
o his scena io exposed ha o - a m ac i i ies ha e a as ole in enhancing he g ow h
o u al economy and also educing he po e y le el (lanjouw, 2001; Weijland, 1999).
Simila ly, Oluwa ayo (2009) explo ed he de e minan s o di e si ica ion using obi
eg ession. Using he coe icien s o gende , household size, po e y s a us and access o
c edi acili y we e ound posi i ely a ec ing he likelihood o di e si ica ion index. Also
a s udy conduc ed by Awoniyi & Salman, (2001) e ealed he le el o non- a m income
a ia ion and also obse ed i s e ec on li ing s anda d o a ming household. Fu he
using logis ic eg ession hey examined he ac o s ha a ec ed he a me s decision o
go o he non- a m income di e si ica ion including age, male headed household, ha ing
o mal educa ion, household po e y s a us and a m size. The impo an judgmen o he
s udy was ha a m households li ing unde he po e y line among hem 53.9 pe cen
we e hose whose household heads we e no be o hed in o - a m ac i i ies o ea n mo e.
Concluding ha he household heads we e mo e suscep ible o po e y i hey had no o -
a m income sou ces when compa ed wi h a ming household ha we e in ol ed in non-
a m income gene a ing sou ces.
As ag icul u e is he spine o Pakis an’s economy, bu is going h ough a dilemma.
One o he majo issues is ins abili y o a m income, especially, o poo a me s. The
inc easing end o is educing due o he bu den o popula ion and di ision o land. F om
o e all a m size nea ly 80% o consis less han 12.5 ac es (GOPun, 2015). Economic
p essu e has o ced many household o go o addi ional income sou ces like o - a m
ac i i y o handle he p oblem o income a iabili y. Nowadays o - a m income is
becoming a majo pa o li elihood s a egies o u al households in ou coun y. O -
a m employmen s a e hose ac i i ies om which a me can ea n apa om a m wo k.
I includes all non-c op ac i i ies ha a e no di ec ly ela ed o c ops.
The main objec i e o he s udy is o de e mine he ac o s in luencing he a ming
household o pa icipa ing in o - a m employmen ac i i ies. I is e y impe a i e o
highligh he en y ba ie s and es ain s o household o in ol e in hese ac i i ies. I
o - a m income can inc ease household income and lessen he isk and inequali y, hen
i is impo an ha such limi a ions a e ul illed. This pape will p o ide addi ional
in o ma ion abou o - a m ac i i y pa icipa ion in u al a eas o Pakis an o accessing
he s a us o o - a m ac i i y pa icipa ion.
Me hodology
Conside ing he a ea whe e co on was cul i a ed in a la ge p opo ion. Punjab
p o ince consis s o nine di isions on adminis a i e basis. Among hese nine di isions
h ee di isions we e selec ed on he basis o as a ea sown unde co on. F om each
di ision one dis ic was selec ed wi h highes sha e o a ea unde co on c op. Mainly
h ee dis ic s namely Bahawalpu , Muza a ga h and Veha i o Punjab p o ince we e
selec ed acco ding o a ea cul i a ed unde co on c op. Acco ding p o incial s a is ics in
Bahawalpu (272,000 hec a es), Muza a ga h (94,000 hec a es) and Veha i (223,000
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
710
hec a es) a ea was cul i a ed (GOPun. 2012). Finally hese dis ic s we e selec ed o
su ey.
Ru al o - a m ac i i ies in he s udy a eas we e ca ego ized in o h ee ypes; i)
Se ices including all ypes o employmen in public and p i a e sec o ins i u ions,
eache s, lawye s, illage doc o s e c. ii) Business en e p ises like shop keeping,
commission agen s and e ilize s o pes icide business holde in g ain ma ke , any ype
o ade , con ac o se ices e c. iii) O - a m labou comp ising di e en ypes o
mechanics, daily labou in u al a eas, anspo ope a ions, cons uc ion labou e c.
The p ima y da a was collec ed om he co on g owe s om he abo e men ioned
dis ic s. Mul is age s age clus e sampling echnique was used o selec 180 esponden s.
F om each dis ic 60 esponden s we e in e iewed andomly consis ing 20 esponden s
om each g oup o o - a m ac i i ies a o emen ioned. The su ey was p emedi a ed o
collec he in o ma ion abou a ming household o e all composi ion and o he socio-
economic cha ac e is ics, which include in o ma ion abou he in ol emen o indi idual
household membe in di e en income gene a ing ac i i ies.
Model
The heo e ical backg ound o in es iga e he e ec s o he elemen s p omp ing he
pa aking in non- a m economic deeds has i s o igins in he h eshold heo y o decision
making. Acco ding o his heo y when he s eng h o he inducemen s goes beyond he
indi iduals eac ion hen an ac ion happens (Hill and kau 1973). The choice o
in ol emen in non- a m economic ac i i ies may ha e wo op ions ei he o join o no
o pa icipa e. Fu he conside ing he pa icipa ion in o - a m ac i i ies as a unc ion o
a m and a me cha ac e is ics i.e. age, educa ion, a m size e c. The dependen a iable
o he model is pa icipa ion in di e en ypes o o - a m ac i i ies such as business,
se ices and o - a m labou . Since he dependen a iable was dicho omous, hence OLS
canno be used. Households ha pa icipa ed in o - a m ac i i ies we e deno ed by 1
o he wise 0 o HHs no pa icipa ing in o a m ac i i ies. As he dependen a iable is
dicho omous, a logis ic eg ession model is pu sued o es ima e he household
pa icipa ion in o - a m ac i i ies (Isgu , 2004; Akaakohol & Aye 2014; Da y & Kuunibe
2012; Lanjouw, 2001).
The logis ic model o his s udy is indica ed as
Pi
(1−Pi)=1+exp(Zi)
1+exp(−Zi)
As he abo e equa ion is non-linea , i can be linea ize by aking he na u al log, hen
he gi en model is
Li=Ln⌈Zi
(1−Pi)⌉Zi
=βo +β1Xi1 +⋯+β9Xi9 +e
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
711
Whe e Pi (1−Pi)⁄ is he a io o he p obabili y ha he a me will be in ol ed in
o - a m ac i i y o he p obabili y ha a me will no in ol e in speci ic o - a m
ac i i y. So he endogenous a iable is bina y and i has wo alues 1 and 0. I a a me
has o - a m ac i i y hen i s alue is 1 and 0 o he a me s who do no ha e pa icula
o - a m wo k.
βo = cons an
β1 – β9 = logis ic eg ession coe icien s
Xi1 = Age o esponden
Xi2 = educa ion (numbe o yea s o o mal educa ion)
Xi3 = a m size ( o al a ming a ea o esponden
Xi4 = loca ion (dis ance om main ci y)
Xi5 = household wo ke s (numbe o ea ning membe in household)
Xi6 = dummy a iable o ha ing access o oad (1 o ha ing access and 0
o he wise)
Xi7 = dependency a io (ob ained by di iding o al membe o household and ea ning
membe )
Xi8 = a ming expe ience o household in yea s
Xi9 = amily size ( o al membe s o amily)
е = e o e m
Resul s and Discussion
The summa y s a is ics o he a iables used in he model a e p esen ed in able-1.
Howe e , able-1 indica es ha he a me s a e on a e age 45 yea s o old. On a e age
a me s had 9 yea s o o mal educa ion and 20 yea s o a ming expe ience. The a e age
a ming a ea is 15 ac e and he mean dis ance o he illage om he main ci y was 11
km; simila ly he mean alue o numbe o ea ning membe s in a amily and amily size
we e 2.74 and 8.25 membe s espec i ely. Majo i y o he a me s om a ge a ea had
access o oad om hei illage. The a e age alue o dependency a io is 3.21 o o al
membe s o household and o al ea ning membe s in a amily espec i ely.
Table 1 Summa y s a is ics o he a iables used o eg ession
Va iables
Mean
SD
Min
Max
Age
45.42
8.18
24.00
68.00
Educa ion (Yea s)
9.61
3.39
0.00
16.00
To al a ming a ea (ac e)
15.23
7.22
3.00
37.00
Loca ion (Dis ance om main ci y in Km)
11.33
4.35
3.00
30.00
Numbe o ea ning membe s in amily
2.74
0.73
2.00
4.00
Access o oad
0.94
0.24
0.00
1.00
Dependency a io
3.21
1.02
1.67
6.00
Fa ming expe ience (Yea s)
20.99
8.82
2.00
45.00
Family size
8.25
1.47
4.00
12.00
Composi ion o A e age Annual Household Income
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
712
Da a ela ed o income ob aining om each c op was collec ed om esponden s.
Income on a m basis was calcula ed o he comple e c opping season.
Table 2 A e age Annual Income om di e en sou ces (PKR)
Income Sou ces
Business
O - a m labou
Se ices
C op
566,517.02
516,861.95
458,969.90
Li es ock
56,275.00
63,760.00
59,624.16
O he s*
3,920.00
14,000.00
30,100.00
Sub-To al (A)
626,712.02
594,621.95
548,694.07
O - a m Income
Business
118,933.33
-
-
O - a m Labou
-
101,040.00
-
Se ices
-
-
102,760.00
Sub-To al (B)
118,933.33
101,040.00
102,160.00
G and To al (A+B)
745,645.35
695,661.95
650,851.07
*I includes he income om pension, emi ances and o he unea ned income.
Table 2 di ulges he sha e o di e en income sou ces o o e all household income. The
o e all household income consis s o c op income, li es ock income and o he income
sou ces i.e. pension, emi ances e c. The o al household income also has con ibu ion
om o - a m income sou ces such as business, o - a m labou and se ices. These
esul s can be coincided wi h Baba unde, Olagunju, Fakayode, and Adejobi (2010) and
De Jan y and Sadoule (2001).
De e minan s o Pa icipa ion in o - a m Ac i i ies
The pa ame e s o he logis ic model es ima ed o iden i y he elemen s p omp ing
pa icipa ion in o - a m ac i i ies a e p esen ed in able 3, 4 and 5.
A logis ic eg ession model was used sepa a ely o he business, o - a m labou and
se ices sec o s o explo e he signi ican ac o s a ec ing he pa icipa ion o a me s in
hese h ee ypes o o - a m ac i i ies.
Table 3 Logis ic eg ession esul s o pa icipa ion in Business ac i i ies
Va iables
β
S.E.
Exp(β)
Age
0.008
0.038
1.008
Educa ion
-0.063
0.059
0.939
To al a ming a ea
**0.05
0.025
1.051
loca ion
0.06
0.041
1.062
Household Wo ke s
**2.814
1.218
16.681
Access o oad (dummy)
***2.431
0.876
0.088
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
713
Dependency a io
***2.872
1.078
17.679
Fa ming expe ience
**-0.084
0.037
0.919
Family Size
**-1.294
0.511
0.274
Cons an
-4.271
3.353
0.014
Model P edic ion success
(MPS)
72.8%
Log-likelihood a io
190.299
Hosme and Lemeshow Tes
(d =8) signi icance es esul 7.635 (p-
alue=0.470)
Cox & Snell R2
0.194
Negelke ke R2
0.270
*Signi icance a 10%; **Signi icance a 5%; ***Signi icance a 1%
Table 3 elucida es he esul s o bina y logis ic eg ession, explaining he de e minan s
o a me s o pa icipa e in o - a m ac i i ies like business. The Hosme and Lemeshow
(H-L) es esul s show non-signi ican indica ing ha he model is good i . While Cox
&Snell R2 and Negelke ke R2 alues a e 0.194 and 0.270 which indica es ha abou 19
o 27 pe cen a ia ions a e explained by he model. The model p edic i e abili y is
72.8%.
The coe icien s o he a iables do no gi e he di ec in o ma ion abou he e ec o
changes in explana o y a iables on he p obabili y o pa icipa ion in business ac i i ies.
To know his odds a io/Exp(β) is necessa y o disce n. This is he a ios o he p obabili y
o pa icipa ing in business ac i i ies o he p obabili y ha he pe son will no ake pa .
The a iables which a e a ec ing signi ican ly he p obabili y o pa icipa ion in business
ac i i ies a e o al a ming a ea, household wo ke s, access o oad, dependency a io,
a ming expe ience and amily size, among which o al a ming a ea, household wo ke s,
access o oad, dependency a io a e he a iables which a e posi i ely changing he
p obabili y. Howe e a ming expe ience and amily size ha e nega i e signs. The esul s
explain ha by inc easing he o al a ming a ea he chances o a me o go o business
ac i i ies inc eases by he alue o associa ed odds a io which is 1.05. The e may be he
eason ha by inc easing a ming a ea household may ea n mo e and so he e is chance
o in es highe income in business ac i i ies. Simila inding a e also epo ed by
(Rahman, 2013).
Simila ly he p obabili y o pa icipa ing in business ac i i ies inc eases wi h ising
numbe o household wo ke s and dependency a io. I is unde s andable ha as
household wo ke s and dependency a io inc eases he p obabili y ha he o he membe s
will go o business ype o ac i i ies ises. On he o he hand, a ming expe ience and
amily size dec eases he p obabili y o business ac i i ies. The a me s which ha e mo e
access o oad hey ha e mo e chance o doing business ac i i ies. These esul s a e
expec ed because ha ing mo e in as uc u e would accele a e he p obabili y o doing
business. As epo ed by Baba unde, Olagunju, Fakayode, and Adejobi (2010) he
coe icien o amily size is unexpec edly nega i e as he e may be eason o ime
cons ain as la ge household may ha e mo e child en o old people in hei amily so
mo e ime is equi ed o hei ca e.
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
714
Table 4 Logis ic eg ession esul s o pa icipa ion in O - a m Labou ac i i ies
Va iables
β
S.E.
Exp(β)
Age
-0.025
0.037
0.976
Edu
***-0.168
0.058
0.845
To al a ming a ea
0.008
0.024
1.008
loca ion
*-0.081
0.045
0.922
Household Wo ke s
-0.821
1.001
0.44
Access o oad (dummy)
1.542
1.106
4.673
Dependency a io
-0.63
0.863
0.532
Fa ming expe ience
**0.061
0.032
1.063
Family Size
0.293
0.386
1.34
Cons an
1.819
3.054
6.163
Model P edic ion Success
(MPS)
72.8%
Log-likelihood a io
200.061
Hosme and Lemeshow Tes
(d =8) signi icance es esul 2.833 (p-
alue=0.944)
Cox & Snell R2
0.149
Negelke ke R2
0.207
*Signi icance a 10%; **Signi icance a 5%; ***Signi icance a 1%
Ano he eg ession was pe o med using he dependen a iable o - a m labou wi h
same independen a iables. The esul s a e p esen ed in able 4. Resul s shows ha he
Hosme and Lemeshow es (H-L) alues a e non-signi ican do model is good i and he
alues o Cox & Snell R2 and Negelke ke R2 indica e ha 14 o 20 pe cen a ia ions
a e explained by he model wi h MPS alue 72.8%. The p obabili y o pa icipa ion in
o - a m labou ac i i ies dec eases wi h educa ion and loca ion. Simila ly Da y &
Kuunibe (2012); Rahman, (2013) also epo ed such ype o indings. Mo e educa ed and
a me loca ed a away om main ci y s and a 0.845 and 0.922 chance o no ge ing he
o - a m labou ac i i ies. Because gene ally mo e educa ed people will no p e e o go
o labou ac i i ies, also i is di icul o he people esiding in emo e a eas o come o
ci y a eas o ind labou oppo uni y. Howe e mo e expe ienced a me s a e mo e likely
o pa icipa e in labou ac i i ies. Usually small land holding a me s y o ind labou
ac i i ies o ge mo e income in addi ion o a m income.
In e na ional Jou nal o Managemen , Accoun ing and Economics
Vol. 2, No. 7, July, 2015
ISSN 2383-2126 (Online)
© IJMAE, All Righ s Rese ed www.ijmae.com
715
Table 5 Logis ic eg ession esul s o pa icipa ion in se ices ac i i ies
Va iables
β
S.E.
Exp(β)
Age
0.041
0.038
1.041
Edu
**0.26
0.069
1.297
To al a ming a ea
**-0.054
0.026
0.948
loca ion
0.015
0.041
1.015
Household Wo ke s
-1.363
1.036
0.256
Access o oad (dummy)
**2.278
1.146
9.757
Dependency a io
**-1.803
0.918
0.165
Fa ming expe ience
-0.005
0.033
0.995
Family Size
*0.727
0.401
2.069
Cons an
-3.116
3.355
0.044
Model P edic ion Success
(MPS)
76.1%
Log-likelihood a io
192.327
Hosme and Lemeshow Tes
(d =8) signi icance es esul 11.474 (p- alue =0.176)
Cox & Snell R2
0.185
Negelke ke R2
0.257
*Signi icance a 10%; **Signi icance a 5%; ***Signi icance a 1%
The las eg ession esul s exposed he ac o s in la ing he a me ’s beha io o adop
he se ices ype o o - a m ac i i ies. Table 5 shows ha he model is good i as
exp essed by he (H-L) es alues and 18 o 25 pe cen a ia ions a e explained by he
model which is ep esen ed by Cox & Snell R2 and Negelke ke R2. The model has
highes MPS alue which is 76.1% as compa ed o o he wo eg ession models gi en. In
he case o se ices educa ion, dummy o oad and amily size played a posi i e and
signi ican ole, while o al a ming a ea and dependency a io we e nega i e. The odd
a io inc eases by 1.297 o educa ed people o go o se ices. Acco ding o (Go don &
G aig, 2001) mo e educa ed household o u al popula ion ha e mo e igh o en y in o
employmen oppo uni ies. Likewise De Jan y & Sadoule , (2001) poin ed ou he
educa ion as one o he main elemen o go o non- a m sec o . Simila ly a me s ha ing
mo e access o oad and la ge amily size a e mo e likely o ake pa in se ices ype o
o - a m ac i i ies. The odds o educa ion, access o oad and amily size inc eases abou
1.297, 9.757 and 2.069 espec i ely. Howe e he odd a io o o al a ming a ea and
dependency a io dec eases by abou 0.948 and 0.165.
Conclusion
