UrbanClimate

Author

University for Development Studies Team

Exploring Household Variability: A Socio-Economic and Demographic Classification of Kumbungu and Savelugu, Ghana

1.0 Introduction

This study examines the heterogeneity of household characteristics in the small towns of Kumbungu and Savelugu, located in northern Ghana. By leveraging a comprehensive dataset, we employ advanced statistical techniques to systematically classify households based on their socio-economic, demographic, and livelihood attributes. The goal is to uncover meaningful household typologies, providing deeper insights into the structure and composition of communities within these towns.

To achieve this, we utilize a combination of Principal Component Analysis (PCA) and Cluster Analysis within the R programming environment. Given the mixed nature of the dataset comprising both numerical and categorical variables PCA serves as a dimensionality reduction technique, transforming complex, high-dimensional data into a set of uncorrelated principal components while preserving key patterns. This step enhances the efficiency and interpretability of the subsequent clustering process, which is applied to group households into distinct categories based on shared characteristics.

By integrating PCA with clustering techniques, we aim to develop a robust household classification system, enabling a clearer understanding of socio-economic disparities, livelihood strategies, and community structures in Kumbungu and Savelugu. The findings from this study will provide valuable insights for policy formulation, urban planning, and targeted socio-economic interventions in small-town settings.

2.0 Description of Study Area

The study area includes Kumbungu (9.55913, -0.94671) and Savelugu (9.62441, -0.8253), both located in the northern region of Ghana. The area is characterized by a tropical savanna climate with distinct wet and dry seasons, which influences land use and land cover (LULC) dynamics.

The dominant land use and land cover (LULC) types include agricultural land, savannah vegetation, settlements, and water bodies. Agriculture, particularly smallholder farming, is the primary land use, with crops such as maize, rice, and groundnuts cultivated. Grasslands and scattered trees dominate the natural vegetation, while urban expansion in Savelugu has led to increased built-up areas. Water bodies, including seasonal streams and small reservoirs, support irrigation and livestock activities.

The rapid land use changes in the area, driven by population growth and agricultural expansion as seen in Figure 1, highlight the need for sustainable land management strategies to balance development with environmental conservation.

Figure 1. Land use Land cover of the project area: Kumbungu (Left) and Savelugu (Right)

3.0 Variable used for Study

This section present the variables that characterize our households used for the analysis. The analysis was divided into demongraphic, land tenure, and income diversification.

Module Variable Used Description
1. Population Characteristics Small town Name of area (Sub-sector)
Ge Gender of household members
Edu Education status of household members
YrsEdu Years of education completed
EduStatus Current education status
Relation with HH Relationship to household head
Child Indicates if the person is a child in the household
Siblings Indicates if the person is a sibling in the household
Spouse Indicates if the person is a spouse in the household
HH head Identifies the household head
Parent Indicates if the person is a parent in the household
Grandparent Indicates if the person is a grandparent in the household
Grandchild Indicates if the person is a grandchild in the household
Other relative Indicates if the person is another type of relative
No family Indicates if the person is not related to the household
2. Land Tenure Arrangements (Inside or Outside Town) LandOwnership Type of land ownership (private, communal, etc.)
HHPreviousArea Previous area of household residence
HHYrsStay Number of years the household has stayed in the area
LandReasforCurrentStay Reason for current land occupation
PreviousLoc Previous location of residence
HHPeriodOfStay Length of stay in the previous location
LandAggrAcc Household agreement on land access
RemitYesCash Household receives cash remittances
RemitYesKind Household receives in-kind remittances
OtherReasonReasforCurrentStay Other reasons for current residence
3. House Tenure (Ownership, Rental, Other) LandAggrePrevious Household’s previous land agreement
4. Timing of Settlement
5. Receiving & Sending Remittances (Translocal Livelihood Indicator) RemitNo Household does not receive remittances
RemSentCshYes Household sends cash remittances
RemitDK Household is unsure about remittance reception
RemSentKindYes Household sends in-kind remittances
RemSentNo Household does not send remittances
RemSentDK Household is unsure about sending remittances
6. Income Diversification HHIncomePrimary Main source of income (e.g., farming, trade, labor)
HHIncomeFarm Income from farming activities
HHIncomeLivestock Income from livestock farming
NonAgLabInc Income from non-agricultural labor
WildProdInc Income from wild products
SheaInc Income from shea nut production
CottonInc Income from cotton production
AgTradeInc Income from agricultural trade
NonAgTradeInc Income from non-agricultural trade
ArtisanInc Income from artisan work (crafts, handmade goods)
PupilTeachInc Income from teaching pupils
TrainedTeachInc Income from professional teaching
ContractorInc Income from contract work
CashIncPension Income from pensions
CashIncGovSupport Income from government support programs
CashIncNGO Income from NGOs
CashIncOther Other sources of income
7. Changes in Income-Generating Activities Over Time TimberPrice5Yr Changes in timber prices over five years
WildFoodPrice5Years Changes in wild food prices over five years
AgriProdPrice5Yr Changes in agricultural product prices over five years
8. Transition from Low to High Income Diversification FuelwoodTimeChange5Y Changes in time spent collecting fuelwood over five years
CharPrice5Yr Changes in charcoal prices over five years
DrinkTimeChange5Years Changes in time spent on drinking water collection over five years

4.0 Descriptive Statistics

4.1 Kumbungu

This section display descriptive analysis of variables used for Kumbungu.


Variable: Ge 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    39       13.0                  13.0
2     1   262       87.0                 100  

-----------------------------------------

Variable: LandOwnership 
# A tibble: 5 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1   149     50.7                    50.7
2     2   121     41.2                    91.8
3     3     2      0.680                  92.5
4     4    14      4.76                   97.3
5     5     8      2.72                  100  

-----------------------------------------

Variable: LandReasforCurrentStay 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   123      97.6                   97.6
2     1     3       2.38                 100  

-----------------------------------------

Variable: PreviousLoc 
# A tibble: 4 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    59      46.8                   46.8
2     2    37      29.4                   76.2
3     3    20      15.9                   92.1
4     4    10       7.94                 100  

-----------------------------------------

Variable: HHPreviousArea 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   174         58                    58
2     1   126         42                   100

-----------------------------------------

Variable: HHPeriodOfStay 
# A tibble: 4 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    22      17.3                   17.3
2     2     6       4.72                  22.0
3     3    92      72.4                   94.5
4     4     7       5.51                 100  

-----------------------------------------

Variable: LandAggrAcc 
# A tibble: 5 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    40      33.1                   33.1
2     2    49      40.5                   73.6
3     3    16      13.2                   86.8
4     4     7       5.79                  92.6
5     5     9       7.44                 100  

-----------------------------------------

Variable: RemitYesCash 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   240       79.7                  79.7
2     1    61       20.3                 100  

-----------------------------------------

Variable: OtherReasonReasforCurrentStay 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   119      94.4                   94.4
2     1     7       5.56                 100  

-----------------------------------------

Variable: LandAggrePrevious 
# A tibble: 4 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1     7      15.2                   15.2
2     2    37      80.4                   95.7
3     4     1       2.17                  97.8
4     5     1       2.17                 100  

-----------------------------------------

Variable: RemitYesKind 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   264       87.7                  87.7
2     1    37       12.3                 100  

-----------------------------------------

Variable: RemitNo 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    84       27.9                  27.9
2     1   217       72.1                 100  

-----------------------------------------

Variable: RemitDK 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   297      98.7                   98.7
2     1     4       1.33                 100  

-----------------------------------------

Variable: RemSentCshYes 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   252       83.7                  83.7
2     1    49       16.3                 100  

-----------------------------------------

Variable: RemSentKindYes 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   240       79.7                  79.7
2     1    61       20.3                 100  

-----------------------------------------

Variable: RemSentNo 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    94       31.2                  31.2
2     1   207       68.8                 100  

-----------------------------------------

Variable: RemSentDK 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   296      98.3                   98.3
2     1     5       1.66                 100  

-----------------------------------------

Variable: HHYrsStay 
# A tibble: 31 × 4
   Value Count Percentage Cumulative_Percentage
   <chr> <int>      <dbl>                 <dbl>
 1 1         3       3.61                  3.61
 2 10        4       4.82                  8.43
 3 11        1       1.20                  9.64
 4 12        3       3.61                 13.3 
 5 13        1       1.20                 14.5 
 6 14        2       2.41                 16.9 
 7 15        4       4.82                 21.7 
 8 16        1       1.20                 22.9 
 9 17        1       1.20                 24.1 
10 18        2       2.41                 26.5 
# ℹ 21 more rows

-----------------------------------------

Variable: HHIncomePrimary 
# A tibble: 17 × 4
   Value Count Percentage Cumulative_Percentage
   <dbl> <int>      <dbl>                 <dbl>
 1     0    10      3.44                   3.44
 2     1   159     54.6                   58.1 
 3     2     9      3.09                  61.2 
 4     4     1      0.344                 61.5 
 5     5     2      0.687                 62.2 
 6     6     2      0.687                 62.9 
 7     8     3      1.03                  63.9 
 8     9     6      2.06                  66.0 
 9    10     7      2.41                  68.4 
10    11    29      9.97                  78.4 
11    12     2      0.687                 79.0 
12    13    21      7.22                  86.3 
13    14     3      1.03                  87.3 
14    15     6      2.06                  89.3 
15    16     1      0.344                 89.7 
16    17     1      0.344                 90.0 
17    18    29      9.97                 100   

-----------------------------------------

Variable: HHIncomeFarm 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   157       52.7                  52.7
2     1   141       47.3                 100  

-----------------------------------------

Variable: HHIncomeLivestock 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   199       66.6                  66.6
2     1   100       33.4                 100  

-----------------------------------------

Variable: NonAgLabInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   275      91.7                   91.7
2     1    25       8.33                 100  

-----------------------------------------

Variable: WildProdInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   291         97                    97
2     1     9          3                   100

-----------------------------------------

Variable: SheaInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   219       72.8                  72.8
2     1    82       27.2                 100  

-----------------------------------------

Variable: CottonInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   300     99.7                    99.7
2     1     1      0.332                 100  

-----------------------------------------

Variable: AgTradeInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   223       74.3                  74.3
2     1    77       25.7                 100  

-----------------------------------------

Variable: NonAgTradeInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   251       83.7                  83.7
2     1    49       16.3                 100  

-----------------------------------------

Variable: ArtisanInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   234       79.1                  79.1
2     1    62       20.9                 100  

-----------------------------------------

Variable: PupilTeachInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   295      98.3                   98.3
2     1     5       1.67                 100  

-----------------------------------------

Variable: TrainedTeachInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   260       86.7                  86.7
2     1    40       13.3                 100  

-----------------------------------------

Variable: ContractorInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   290      96.3                   96.3
2     1    11       3.65                 100  

-----------------------------------------

Variable: CashIncPension 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   288         96                    96
2     1    12          4                   100

-----------------------------------------

Variable: CashIncGovSupport 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   293      98.7                   98.7
2     1     4       1.35                 100  

-----------------------------------------

Variable: CashIncNGO 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   297         99                    99
2     1     3          1                   100

-----------------------------------------

Variable: CashIncOther 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   268      90.5                   90.5
2     1    28       9.46                 100  

-----------------------------------------

Variable: AccWater 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0     7       2.33                  2.33
2     1   294      97.7                 100   

-----------------------------------------

Variable: ToiletStatus 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   212       70.4                  70.4
2     1    89       29.6                 100  

-----------------------------------------

Variable: HealthUseYes 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    14       4.65                  4.65
2     1   287      95.3                 100   

-----------------------------------------

Variable: SanNeedSW 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   162       54.9                  54.9
2     1   133       45.1                 100  

-----------------------------------------

Variable: NatResSpiritualUse 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   278      95.2                   95.2
2     1    14       4.79                 100  

-----------------------------------------

Variable: LandAccCurrent 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   172       57.9                  57.9
2     1   125       42.1                 100  

-----------------------------------------

Variable: WaterReAccLtd 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   125     99.2                    99.2
2     1     1      0.794                 100  

-----------------------------------------

Variable: AssetCompu 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   238       79.9                  79.9
2     1    60       20.1                 100  

-----------------------------------------

Variable: AssetTele 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0     3      0.997                 0.997
2     1   298     99.0                 100    

-----------------------------------------

Variable: AssetTV 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    48         16                    16
2     1   252         84                   100

-----------------------------------------

Variable: LandReAccLtd 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   122      96.8                   96.8
2     1     4       3.17                 100  

-----------------------------------------

Variable: Edu 
# A tibble: 9 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    14      4.67                   4.67
2     2     4      1.33                   6   
3     3    22      7.33                  13.3 
4     4     9      3                     16.3 
5     5    36     12                     28.3 
6     6     4      1.33                  29.7 
7     7    52     17.3                   47   
8     9   158     52.7                   99.7 
9    10     1      0.333                100   

-----------------------------------------

Variable: YrsEdu 
# A tibble: 20 × 4
   Value Count Percentage Cumulative_Percentage
   <dbl> <int>      <dbl>                 <dbl>
 1     0   157     64.1                    64.1
 2     3     3      1.22                   65.3
 3     4     1      0.408                  65.7
 4     5     2      0.816                  66.5
 5     6     5      2.04                   68.6
 6     7     6      2.45                   71.0
 7     8     5      2.04                   73.1
 8     9     4      1.63                   74.7
 9    10     4      1.63                   76.3
10    11     7      2.86                   79.2
11    12     9      3.67                   82.9
12    13     1      0.408                  83.3
13    14     6      2.45                   85.7
14    15    10      4.08                   89.8
15    16    10      4.08                   93.9
16    17     4      1.63                   95.5
17    18     2      0.816                  96.3
18    19     7      2.86                   99.2
19    20     1      0.408                  99.6
20    21     1      0.408                 100  

-----------------------------------------

Variable: AccElec 
# A tibble: 3 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    14      4.65                   4.65
2     1   286     95.0                   99.7 
3     2     1      0.332                100   

-----------------------------------------

Variable: CookingPlace 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    17       5.65                  5.65
2     1   284      94.4                 100   

-----------------------------------------

Variable: FoodEnerType 
# A tibble: 3 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0     7       2.33                  2.33
2     1   275      91.4                  93.7 
3     2    19       6.31                100   

-----------------------------------------

Variable: AssetRadio 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    64       21.3                  21.3
2     1   237       78.7                 100  

-----------------------------------------

Variable: AssetRefrig 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   156       51.8                  51.8
2     1   145       48.2                 100  

-----------------------------------------

4.2 Savelugu

This section display descriptive analysis of variables used for Savelugu.


Variable: Ge 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    40       12.8                  12.8
2     1   272       87.2                 100  

-----------------------------------------

Variable: LandOwnership 
# A tibble: 3 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1   181      62.8                   62.8
2     2   104      36.1                   99.0
3     3     3       1.04                 100  

-----------------------------------------

Variable: LandReasforCurrentStay 
# A tibble: 1 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   312        100                   100

-----------------------------------------

Variable: PreviousLoc 
# A tibble: 4 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    89      60.5                   60.5
2     2    27      18.4                   78.9
3     3    20      13.6                   92.5
4     4    11       7.48                 100  

-----------------------------------------

Variable: HHPreviousArea 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   160       51.8                  51.8
2     1   149       48.2                 100  

-----------------------------------------

Variable: LandAggrAcc 
# A tibble: 5 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    43      33.1                   33.1
2     2    47      36.2                   69.2
3     3    18      13.8                   83.1
4     4     3       2.31                  85.4
5     5    19      14.6                  100  

-----------------------------------------

Variable: RemitYesCash 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   253       81.1                  81.1
2     1    59       18.9                 100  

-----------------------------------------

Variable: OtherReasonReasforCurrentStay 
# A tibble: 1 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   312        100                   100

-----------------------------------------

Variable: LandAggrePrevious 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   309     99.0                    99.0
2     1     3      0.962                 100  

-----------------------------------------

Variable: RemitYesKind 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   284      91.0                   91.0
2     1    28       8.97                 100  

-----------------------------------------

Variable: RemitNo 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    78         25                    25
2     1   234         75                   100

-----------------------------------------

Variable: HHYrsOcc 
# A tibble: 4 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    58      38.9                   38.9
2     2    11       7.38                  46.3
3     3    73      49.0                   95.3
4     4     7       4.70                 100  

-----------------------------------------

Variable: HHPreviousStay 
# A tibble: 3 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    36      62.1                   62.1
2     1    18      31.0                   93.1
3     2     4       6.90                 100  

-----------------------------------------

Variable: RemitDK 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   305      97.8                   97.8
2     1     7       2.24                 100  

-----------------------------------------

Variable: RemSentCshYes 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   268       85.9                  85.9
2     1    44       14.1                 100  

-----------------------------------------

Variable: RemSentKindYes 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   250       80.1                  80.1
2     1    62       19.9                 100  

-----------------------------------------

Variable: RemSentNo 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    96       30.8                  30.8
2     1   216       69.2                 100  

-----------------------------------------

Variable: RemSentDK 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   300      96.2                   96.2
2     1    12       3.85                 100  

-----------------------------------------

Variable: Edu 
# A tibble: 7 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1    16      5.21                   5.21
2     3    27      8.79                  14.0 
3     4    21      6.84                  20.8 
4     5    25      8.14                  29.0 
5     7    54     17.6                   46.6 
6     8     1      0.326                 46.9 
7     9   163     53.1                  100   

-----------------------------------------

Variable: YrsEdu 
# A tibble: 19 × 4
   Value Count Percentage Cumulative_Percentage
   <dbl> <int>      <dbl>                 <dbl>
 1     0   162       64.8                  64.8
 2     1     1        0.4                  65.2
 3     4     2        0.8                  66  
 4     5     6        2.4                  68.4
 5     6     6        2.4                  70.8
 6     7     2        0.8                  71.6
 7     8     2        0.8                  72.4
 8     9     4        1.6                  74  
 9    10     3        1.2                  75.2
10    11     3        1.2                  76.4
11    12    15        6                    82.4
12    13     1        0.4                  82.8
13    15    10        4                    86.8
14    16    11        4.4                  91.2
15    17     5        2                    93.2
16    19     4        1.6                  94.8
17    20    10        4                    98.8
18    25     2        0.8                  99.6
19    30     1        0.4                 100  

-----------------------------------------

Variable: AccElec 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    38       12.2                  12.2
2     1   274       87.8                 100  

-----------------------------------------

Variable: CookingPlace 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    20       6.41                  6.41
2     1   292      93.6                 100   

-----------------------------------------

Variable: FoodEnerType 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0    10       3.44                  3.44
2     1   281      96.6                 100   

-----------------------------------------

Variable: HHIncomePrimary 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     1   145       46.8                  46.8
2     2   165       53.2                 100  

-----------------------------------------

Variable: HHIncomeFarm 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   168       54.0                  54.0
2     1   143       46.0                 100  

-----------------------------------------

Variable: HHIncomeLivestock 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   226       72.7                  72.7
2     1    85       27.3                 100  

-----------------------------------------

Variable: HHIncomeAgLab 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   287      92.3                   92.3
2     1    24       7.72                 100  

-----------------------------------------

Variable: NonAgLabInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   304      97.7                   97.7
2     1     7       2.25                 100  

-----------------------------------------

Variable: WildProdInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   309     99.4                    99.4
2     1     2      0.643                 100  

-----------------------------------------

Variable: SheaInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   240       76.9                  76.9
2     1    72       23.1                 100  

-----------------------------------------

Variable: CottonInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   311     99.7                    99.7
2     1     1      0.321                 100  

-----------------------------------------

Variable: AgTradeInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   245       79.3                  79.3
2     1    64       20.7                 100  

-----------------------------------------

Variable: NonAgTradeInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   263       84.3                  84.3
2     1    49       15.7                 100  

-----------------------------------------

Variable: ArtisanInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   243       78.1                  78.1
2     1    68       21.9                 100  

-----------------------------------------

Variable: PupilTeachInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   309     99.0                    99.0
2     1     3      0.962                 100  

-----------------------------------------

Variable: TrainedTeachInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   263       85.1                  85.1
2     1    46       14.9                 100  

-----------------------------------------

Variable: ContractorInc 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   302      97.4                   97.4
2     1     8       2.58                 100  

-----------------------------------------

Variable: CashIncPension 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   308      98.7                   98.7
2     1     4       1.28                 100  

-----------------------------------------

Variable: CashIncGovSupport 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   308     99.0                    99.0
2     1     3      0.965                 100  

-----------------------------------------

Variable: CashIncNGO 
# A tibble: 1 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   310        100                   100

-----------------------------------------

Variable: CashIncOther 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   232       76.8                  76.8
2     1    70       23.2                 100  

-----------------------------------------

Variable: KindIncHH 
# A tibble: 2 × 4
  Value Count Percentage Cumulative_Percentage
  <dbl> <int>      <dbl>                 <dbl>
1     0   256       84.8                  84.8
2     1    46       15.2                 100  

-----------------------------------------

5.0 Demongraphics

5.1 Kumbungu

This section describes dependency ratio by gender and level of education in Kumbungu.This analysis categorizes individuals into two groups: “Old” and “Young,” based on their age, with male (1) and female (0) data represented separately. The table shows the count and percentage of males and females within each age group, with a higher percentage of males in both the “Old” and “Young” categories compared to females.

# A tibble: 4 × 4
  DependencyCategory    Ge Count Percentage
  <chr>              <dbl> <int>      <dbl>
1 Old                    0    22       8.30
2 Old                    1    94      35.5 
3 Young                  0    13       4.91
4 Young                  1   136      51.3 

5.2 Savelugu

This section describes dependency ratio by gender and level of education in Savelugu. This analysis examines the distribution of the “DependencyCategory” variable, categorizing individuals into “Old” and “Young” groups, with binary values indicating the presence or absence of a specific condition. The table presents the count and percentage of each category, showing that the “Young” group with the condition (1 = male) comprises the largest proportion at 52.36%, while the “Old” group without the condition (0 = female) has the smallest proportion at 7.27%.

# A tibble: 4 × 4
  DependencyCategory    Ge Count Percentage
  <chr>              <dbl> <int>      <dbl>
1 Old                    0    20       7.27
2 Old                    1    97      35.3 
3 Young                  0    14       5.09
4 Young                  1   144      52.4 

6.0 Land tenure and Remittance

The “Land Tenure and Remittance” section of the Urban Climate project examines the relationship between land ownership and the financial support received through remittances. It explores how land tenure security influences household livelihoods, access to resources, and the role of remittances in improving economic stability, particularly in urban settings.

6.1 Kumbungu (PCA)

The analysis performed uses Principal Component Analysis (PCA) to reduce the dimensionality of socio-economic and land tenure data while preserving as much variance as possible. The PCA summary reveals that the first three principal components (PCs) explain over 35% of the total variance in the data, with the first component alone accounting for 14.6% of the variance. Variables like “LandOwnership,” “RemitYesCash,” and “RemitYesKind” contribute most to the first principal component, indicating they are highly influential in explaining the variation in the dataset. The subsequent components explain diminishing amounts of variance, with the cumulative variance reaching nearly 100% after 17 dimensions, showing that the most significant relationships in the data are captured by the first few components.

This PCA analysis is particularly relevant to the Urban Climate Project, as it provides insights into the key factors influencing land tenure and remittance behaviors in Kumbungu, which are critical in understanding urban climate challenges. For example, “RemitYesCash” and “RemitYesKind” both have high contributions in the first component, suggesting that remittance flows (both in cash and kind) play an important role in urban adaptation strategies. These factors may influence how individuals or households respond to urban climate risks, such as flooding or heatwaves, as they reflect not only economic activity but also mobility and settlement patterns. By analyzing how these variables interact, the project can identify targeted interventions for climate resilience in urban areas.


Call:
PCA(X = data_pca, scale.unit = TRUE, graph = FALSE) 


Eigenvalues
                       Dim.1   Dim.2   Dim.3   Dim.4   Dim.5   Dim.6   Dim.7
Variance               2.484   1.867   1.625   1.323   1.221   1.132   1.032
% of var.             14.612  10.981   9.557   7.781   7.181   6.661   6.070
Cumulative % of var.  14.612  25.593  35.150  42.932  50.112  56.773  62.843
                       Dim.8   Dim.9  Dim.10  Dim.11  Dim.12  Dim.13  Dim.14
Variance               0.990   0.958   0.851   0.797   0.719   0.671   0.645
% of var.              5.822   5.633   5.003   4.690   4.230   3.949   3.794
Cumulative % of var.  68.665  74.298  79.301  83.991  88.221  92.170  95.964
                      Dim.15  Dim.16  Dim.17
Variance               0.547   0.082   0.058
% of var.              3.215   0.480   0.340
Cumulative % of var.  99.179  99.660 100.000

Individuals (the 10 first)
                                  Dist    Dim.1    ctr   cos2    Dim.2    ctr
1                             |  1.843 | -1.224  0.200  0.441 | -0.331  0.020
2                             |  1.731 | -1.278  0.219  0.545 | -0.306  0.017
3                             |  1.742 | -1.285  0.221  0.544 | -0.304  0.016
4                             |  1.602 | -1.220  0.199  0.580 | -0.034  0.000
5                             |  4.212 | -1.744  0.407  0.171 | -0.139  0.003
6                             |  1.602 | -1.220  0.199  0.580 | -0.034  0.000
7                             |  2.906 |  0.882  0.104  0.092 | -1.895  0.639
8                             |  2.906 |  0.882  0.104  0.092 | -1.895  0.639
9                             |  2.635 | -1.471  0.289  0.312 |  0.029  0.000
10                            |  3.459 | -1.207  0.195  0.122 | -0.376  0.025
                                cos2    Dim.3    ctr   cos2  
1                              0.032 | -0.520  0.055  0.080 |
2                              0.031 | -0.203  0.008  0.014 |
3                              0.031 | -0.195  0.008  0.012 |
4                              0.000 |  0.191  0.007  0.014 |
5                              0.001 |  0.784  0.126  0.035 |
6                              0.000 |  0.191  0.007  0.014 |
7                              0.425 |  0.577  0.068  0.039 |
8                              0.425 |  0.577  0.068  0.039 |
9                              0.000 | -0.007  0.000  0.000 |
10                             0.012 | -2.098  0.900  0.368 |

Variables (the 10 first)
                                 Dim.1    ctr   cos2    Dim.2    ctr   cos2  
LandOwnership                 |  0.083  0.278  0.007 |  0.337  6.096  0.114 |
LandReasforCurrentStay        | -0.129  0.671  0.017 | -0.025  0.033  0.001 |
PreviousLoc                   |  0.061  0.148  0.004 |  0.155  1.283  0.024 |
HHPreviousArea                | -0.066  0.177  0.004 |  0.122  0.791  0.015 |
HHPeriodOfStay                |  0.137  0.751  0.019 | -0.053  0.153  0.003 |
LandAggrAcc                   |  0.052  0.110  0.003 | -0.016  0.013  0.000 |
RemitYesCash                  |  0.635 16.218  0.403 | -0.514 14.129  0.264 |
OtherReasonReasforCurrentStay |  0.115  0.532  0.013 |  0.007  0.003  0.000 |
LandAggrePrevious             | -0.005  0.001  0.000 | -0.084  0.374  0.007 |
RemitYesKind                  |  0.527 11.167  0.277 | -0.442 10.468  0.195 |
                               Dim.3    ctr   cos2  
LandOwnership                  0.456 12.778  0.208 |
LandReasforCurrentStay        -0.161  1.589  0.026 |
PreviousLoc                    0.558 19.132  0.311 |
HHPreviousArea                -0.036  0.078  0.001 |
HHPeriodOfStay                -0.641 25.300  0.411 |
LandAggrAcc                   -0.054  0.178  0.003 |
RemitYesCash                   0.113  0.785  0.013 |
OtherReasonReasforCurrentStay  0.049  0.146  0.002 |
LandAggrePrevious             -0.321  6.331  0.103 |
RemitYesKind                   0.160  1.582  0.026 |

6.2 Kumbungu (Clustering)

Optimal Number of Clusters

The Elbow method and Silhouette method are used to determine the optimal number of clusters. Both methods are visualized using fviz_nbclust, helping to identify the best number of clusters based on within-cluster sum of squares (WSS) and silhouette scores.

K-means Clustering

K-means clustering is performed with 3 clusters, and the clustering results are visualized before and after applying the K-means algorithm using Principal Component Analysis (PCA) and cluster visualization techniques.

6.3 Savelugu (PCA)

The Principal Component Analysis (PCA) results presented here (Savelugu) focus on the dimensional reduction of multiple socio-economic variables relevant to the Urban Climate project. The PCA was performed on a dataset with 17 dimensions, where the first three principal components (PCs) account for over 41% of the variance in the data. Specifically, the first component (Dim.1) explains 18.29% of the variance, the second (Dim.2) accounts for 12.11%, and the third (Dim.3) explains 11.05%. Cumulatively, these first three components capture 41.44% of the total variance, offering a concise summary of the data. The eigenvalues for each dimension indicate how much variance each component explains, with Dim.1 being the most significant. The analysis of the variables also reveals that “RemitYesCash” (remittance cash received) strongly correlates with Dim.1, contributing 16.34% of the variance, suggesting that remittances are an essential variable in understanding socio-economic factors in urban areas.

In the context of the Urban Climate project, this PCA analysis helps identify the key socio-economic factors that may influence urban climate vulnerability, particularly in areas related to land tenure and remittance patterns. For instance, “RemitNo” (no remittance received) significantly contributes to Dim.1, with a high contribution of 22.34%, highlighting the importance of remittances in shaping household economic dynamics. Other factors such as “LandAggrAcc” (land aggreement access) and “HHPreviousArea” (previous household area) also contribute to the clustering of observations, shedding light on how mobility and land ownership impact climate adaptation strategies. The cumulative percentage of variance (up to 84.7%) indicates that a substantial proportion of the data variability can be explained through these principal components, which is crucial for understanding the relationship between socio-economic factors and urban climate resilience.


Call:
PCA(X = data_pca, scale.unit = TRUE, graph = FALSE) 


Eigenvalues
                       Dim.1   Dim.2   Dim.3   Dim.4   Dim.5   Dim.6   Dim.7
Variance               2.743   1.816   1.657   1.495   1.237   1.093   0.986
% of var.             18.286  12.105  11.050   9.966   8.246   7.290   6.576
Cumulative % of var.  18.286  30.392  41.441  51.408  59.654  66.943  73.520
                       Dim.8   Dim.9  Dim.10  Dim.11  Dim.12  Dim.13  Dim.14
Variance               0.894   0.782   0.665   0.616   0.512   0.377   0.072
% of var.              5.962   5.214   4.434   4.110   3.410   2.515   0.481
Cumulative % of var.  79.482  84.696  89.131  93.241  96.651  99.166  99.647
                      Dim.15  Dim.16  Dim.17
Variance               0.053   0.000   0.000
% of var.              0.353   0.000   0.000
Cumulative % of var. 100.000 100.000 100.000

Individuals (the 10 first)
                                  Dist    Dim.1    ctr   cos2    Dim.2    ctr
1                             |  2.321 | -1.234  0.178  0.283 | -0.240  0.010
2                             |  2.102 | -1.295  0.196  0.380 |  0.372  0.024
3                             |  3.079 | -1.314  0.202  0.182 |  0.370  0.024
4                             |  1.775 | -1.264  0.187  0.507 |  0.016  0.000
5                             |  2.036 | -1.207  0.170  0.351 | -0.418  0.031
6                             |  3.459 | -1.140  0.152  0.109 | -0.572  0.058
7                             |  2.064 | -1.200  0.168  0.338 | -0.476  0.040
8                             |  2.691 | -1.344  0.211  0.249 |  0.626  0.069
9                             |  4.691 | -1.139  0.152  0.059 |  0.367  0.024
10                            |  1.775 | -1.264  0.187  0.507 |  0.016  0.000
                                cos2    Dim.3    ctr   cos2  
1                              0.011 | -0.282  0.015  0.015 |
2                              0.031 | -0.442  0.038  0.044 |
3                              0.014 | -0.871  0.147  0.080 |
4                              0.000 |  0.098  0.002  0.003 |
5                              0.042 |  0.712  0.098  0.122 |
6                              0.027 |  0.269  0.014  0.006 |
7                              0.053 |  0.626  0.076  0.092 |
8                              0.054 | -0.491  0.047  0.033 |
9                              0.006 | -0.664  0.085  0.020 |
10                             0.000 |  0.098  0.002  0.003 |

Variables (the 10 first)
                                 Dim.1    ctr   cos2    Dim.2    ctr   cos2  
LandOwnership                 |  0.046  0.078  0.002 | -0.285  4.470  0.081 |
LandReasforCurrentStay        |  0.000  0.000    Inf |  0.000  0.000    Inf |
PreviousLoc                   |  0.041  0.061  0.002 |  0.011  0.007  0.000 |
HHPreviousArea                |  0.003  0.000  0.000 |  0.211  2.442  0.044 |
LandAggrAcc                   | -0.034  0.041  0.001 |  0.231  2.937  0.053 |
RemitYesCash                  |  0.669 16.338  0.448 | -0.178  1.740  0.032 |
OtherReasonReasforCurrentStay |  0.000  0.000    Inf |  0.000  0.000    Inf |
LandAggrePrevious             | -0.123  0.548  0.015 |  0.065  0.234  0.004 |
RemitYesKind                  |  0.488  8.681  0.238 | -0.117  0.760  0.014 |
RemitNo                       | -0.783 22.344  0.613 | -0.124  0.849  0.015 |
                               Dim.3    ctr   cos2  
LandOwnership                  0.385  8.937  0.148 |
LandReasforCurrentStay         0.000  0.000    Inf |
PreviousLoc                   -0.018  0.019  0.000 |
HHPreviousArea                -0.349  7.345  0.122 |
LandAggrAcc                    0.327  6.462  0.107 |
RemitYesCash                   0.437 11.501  0.191 |
OtherReasonReasforCurrentStay  0.000  0.000    NaN |
LandAggrePrevious             -0.037  0.084  0.001 |
RemitYesKind                   0.415 10.412  0.173 |
RemitNo                       -0.500 15.096  0.250 |

6.4 Savelugu (Clustering)

The ideal quantity of clusters
The ideal number of clusters is found using the Elbow and Silhouette methods. In order to determine the optimal number of clusters based on within-cluster sum of squares (WSS) and silhouette scores, both approaches are visualized using fviz_nbclust.

Clustering with K-means
Three clusters are used for K-means clustering, and the clustering outcomes are shown using Principal Component Analysis (PCA) and cluster visualization techniques both before and after the K-means algorithm is applied.

7.0 Income Diversification

Income diversification in the Urban Climate project refers to the various strategies households use to generate income from multiple sources, reducing reliance on a single economic activity. This helps improve resilience to climate-related impacts by ensuring financial stability through activities such as agriculture, trade, livestock, and others.

7.1 Kumbungu PCA

The analysis presented is based on Principal Component Analysis (PCA) applied to a dataset related to urban climate factors and income diversification. PCA is a dimensionality reduction technique that identifies patterns in multivariate data, allowing for the identification of the key variables that explain the most variance. In this case, the PCA has identified several principal components (PCs) that capture the variance in income sources, with Dim.1 explaining 11.83% of the variance, Dim.2 explaining 8.23%, and so on, providing a cumulative 63.94% variance explained by the first nine dimensions. The cumulative percentage of variance is a key indicator of how well the data has been reduced while retaining important information.

The eigenvalues associated with each principal component (e.g., Dim.1 = 2.130, Dim.2 = 1.480) indicate the variance each dimension accounts for. For example, Dim.1 represents the most significant source of variation in income-related factors, followed by Dim.2 and Dim.3. These components reflect how different income sources like “HHIncomeFarm,” “HHIncomePrimary,” and “SheaInc” are related to urban climate factors, with varying degrees of contribution to the total variability. The loadings (cos2 values) for each variable on these components, such as “HHIncomePrimary” having a cos2 of 0.365 for Dim.1, suggest that primary income from households contributes significantly to this principal component.

Finally, analyzing the first few individuals and their projections on the principal components reveals how specific cases or observations from the dataset relate to the broader trends identified in the PCA. For example, individuals with higher projections in Dim.1 and Dim.2 tend to have stronger links with income from agriculture (e.g., “HHIncomeFarm”) and primary household income, aligning with income diversification strategies. The results from this PCA can guide further analysis of the economic resilience of urban populations, particularly how income diversification across sectors can mitigate climate risks and enhance urban climate adaptation strategies.


Call:
PCA(X = data_pca, scale.unit = TRUE, graph = FALSE) 


Eigenvalues
                       Dim.1   Dim.2   Dim.3   Dim.4   Dim.5   Dim.6   Dim.7
Variance               2.130   1.480   1.275   1.215   1.190   1.097   1.068
% of var.             11.833   8.225   7.085   6.752   6.612   6.095   5.933
Cumulative % of var.  11.833  20.058  27.142  33.894  40.506  46.601  52.533
                       Dim.8   Dim.9  Dim.10  Dim.11  Dim.12  Dim.13  Dim.14
Variance               1.043   1.009   0.907   0.881   0.836   0.786   0.716
% of var.              5.795   5.607   5.040   4.893   4.646   4.366   3.977
Cumulative % of var.  58.329  63.936  68.977  73.869  78.515  82.881  86.858
                      Dim.15  Dim.16  Dim.17  Dim.18  Dim.19
Variance               0.688   0.664   0.554   0.459   0.000
% of var.              3.821   3.691   3.079   2.550   0.000
Cumulative % of var.  90.680  94.371  97.450 100.000 100.000

Individuals (the 10 first)
                      Dist    Dim.1    ctr   cos2    Dim.2    ctr   cos2  
1                 |  2.715 | -1.425  0.305  0.275 | -0.929  0.187  0.117 |
2                 |  4.156 | -0.171  0.004  0.002 |  1.241  0.334  0.089 |
3                 |  9.794 | -0.573  0.049  0.003 |  4.329  4.058  0.195 |
4                 |  3.591 |  1.218  0.223  0.115 |  1.227  0.326  0.117 |
5                 |  3.063 |  1.660  0.415  0.294 | -0.345  0.026  0.013 |
6                 |  2.803 | -0.835  0.105  0.089 |  0.486  0.051  0.030 |
7                 |  4.704 |  2.742  1.132  0.340 | -0.793  0.136  0.028 |
8                 |  4.182 |  1.389  0.290  0.110 |  0.657  0.093  0.025 |
9                 |  2.063 | -1.740  0.456  0.711 | -0.093  0.002  0.002 |
10                |  7.763 |  0.513  0.040  0.004 |  3.151  2.150  0.165 |
                   Dim.3    ctr   cos2  
1                  0.746  0.140  0.076 |
2                 -0.184  0.008  0.002 |
3                  0.165  0.007  0.000 |
4                 -1.000  0.251  0.078 |
5                 -0.441  0.049  0.021 |
6                 -0.425  0.045  0.023 |
7                 -0.613  0.095  0.017 |
8                 -0.258  0.017  0.004 |
9                 -0.357  0.032  0.030 |
10                 0.440  0.049  0.003 |

Variables (the 10 first)
                     Dim.1    ctr   cos2    Dim.2    ctr   cos2    Dim.3    ctr
HHIncomePrimary   |  0.604 17.140  0.365 | -0.343  7.964  0.118 | -0.083  0.541
HHIncomeFarm      |  0.642 19.376  0.413 | -0.285  5.470  0.081 | -0.023  0.040
HHIncomeLivestock |  0.375  6.604  0.141 |  0.200  2.700  0.040 | -0.096  0.723
HHIncomeAgLab     |  0.499 11.686  0.249 | -0.198  2.646  0.039 |  0.057  0.259
NonAgLabInc       |  0.063  0.187  0.004 |  0.086  0.500  0.007 |  0.233  4.264
WildProdInc       |  0.032  0.049  0.001 | -0.105  0.749  0.011 | -0.140  1.541
SheaInc           |  0.564 14.958  0.319 |  0.152  1.554  0.023 |  0.147  1.702
CottonInc         |  0.067  0.212  0.005 |  0.031  0.065  0.001 | -0.062  0.301
AgTradeInc        |  0.532 13.313  0.284 |  0.284  5.450  0.081 | -0.031  0.075
NonAgTradeInc     |  0.141  0.938  0.020 | -0.078  0.413  0.006 |  0.331  8.580
                    cos2  
HHIncomePrimary    0.007 |
HHIncomeFarm       0.001 |
HHIncomeLivestock  0.009 |
HHIncomeAgLab      0.003 |
NonAgLabInc        0.054 |
WildProdInc        0.020 |
SheaInc            0.022 |
CottonInc          0.004 |
AgTradeInc         0.001 |
NonAgTradeInc      0.109 |

7.2 Kumbungu (Clustering)

This section show cluster analysis on income diversification in Kumbungu.

7.3 Savelugu (PCA)

The Principal Component Analysis (PCA) conducted on the Urban Climate project dataset reveals the underlying patterns and relationships among various income sources in the community (Savelugu), such as household income from agriculture, livestock, and non-agricultural activities. The first three principal components (Dim.1, Dim.2, and Dim.3) explain 27.14%, 20.06%, and 19.21% of the variance, respectively, with the cumulative variance for the first three components accounting for 66.41% of the total variation. This indicates that a substantial portion of the income diversification data is captured by these three components, highlighting key variables like HHIncomePrimary, HHIncomeFarm, and HHIncomeAgLab.

For instance, the analysis shows that household income from agriculture (HHIncomeFarm) and primary income sources (HHIncomePrimary) are highly influential in the first component (Dim.1), contributing 19.38% and 17.14%, respectively, to the variance in this dimension. These two variables exhibit substantial correlations with Dim.1, suggesting that income from these sectors is a major factor in shaping the urban climate context in the region. Additionally, non-agricultural income (NonAgLabInc) and wild product income (WildProdInc) appear to have lower weightings in the first dimension but gain more prominence in subsequent components, particularly Dim.3, which is explained mostly by income from non-agricultural trade (NonAgTradeInc).

The PCA results provide a multidimensional view of income diversification, indicating that households in the urban areas depend heavily on agricultural sources, especially from farming and primary activities. The variables showing higher contributions to the first two principal components, such as HHIncomeFarm and HHIncomePrimary, may suggest the importance of agricultural sustainability and resilience in the context of climate adaptation strategies. However, the significant presence of non-agricultural sources, particularly income from trade (AgTradeInc and NonAgTradeInc), in Dim.3 points to the growing economic diversification in response to climate pressures, where households increasingly rely on alternative income streams to buffer against environmental uncertainties.


Call:
PCA(X = data_pca, scale.unit = TRUE, graph = FALSE) 


Eigenvalues
                       Dim.1   Dim.2   Dim.3   Dim.4   Dim.5   Dim.6   Dim.7
Variance               2.130   1.480   1.275   1.215   1.190   1.097   1.068
% of var.             11.833   8.225   7.085   6.752   6.612   6.095   5.933
Cumulative % of var.  11.833  20.058  27.142  33.894  40.506  46.601  52.533
                       Dim.8   Dim.9  Dim.10  Dim.11  Dim.12  Dim.13  Dim.14
Variance               1.043   1.009   0.907   0.881   0.836   0.786   0.716
% of var.              5.795   5.607   5.040   4.893   4.646   4.366   3.977
Cumulative % of var.  58.329  63.936  68.977  73.869  78.515  82.881  86.858
                      Dim.15  Dim.16  Dim.17  Dim.18  Dim.19
Variance               0.688   0.664   0.554   0.459   0.000
% of var.              3.821   3.691   3.079   2.550   0.000
Cumulative % of var.  90.680  94.371  97.450 100.000 100.000

Individuals (the 10 first)
                      Dist    Dim.1    ctr   cos2    Dim.2    ctr   cos2  
1                 |  2.715 | -1.425  0.305  0.275 | -0.929  0.187  0.117 |
2                 |  4.156 | -0.171  0.004  0.002 |  1.241  0.334  0.089 |
3                 |  9.794 | -0.573  0.049  0.003 |  4.329  4.058  0.195 |
4                 |  3.591 |  1.218  0.223  0.115 |  1.227  0.326  0.117 |
5                 |  3.063 |  1.660  0.415  0.294 | -0.345  0.026  0.013 |
6                 |  2.803 | -0.835  0.105  0.089 |  0.486  0.051  0.030 |
7                 |  4.704 |  2.742  1.132  0.340 | -0.793  0.136  0.028 |
8                 |  4.182 |  1.389  0.290  0.110 |  0.657  0.093  0.025 |
9                 |  2.063 | -1.740  0.456  0.711 | -0.093  0.002  0.002 |
10                |  7.763 |  0.513  0.040  0.004 |  3.151  2.150  0.165 |
                   Dim.3    ctr   cos2  
1                  0.746  0.140  0.076 |
2                 -0.184  0.008  0.002 |
3                  0.165  0.007  0.000 |
4                 -1.000  0.251  0.078 |
5                 -0.441  0.049  0.021 |
6                 -0.425  0.045  0.023 |
7                 -0.613  0.095  0.017 |
8                 -0.258  0.017  0.004 |
9                 -0.357  0.032  0.030 |
10                 0.440  0.049  0.003 |

Variables (the 10 first)
                     Dim.1    ctr   cos2    Dim.2    ctr   cos2    Dim.3    ctr
HHIncomePrimary   |  0.604 17.140  0.365 | -0.343  7.964  0.118 | -0.083  0.541
HHIncomeFarm      |  0.642 19.376  0.413 | -0.285  5.470  0.081 | -0.023  0.040
HHIncomeLivestock |  0.375  6.604  0.141 |  0.200  2.700  0.040 | -0.096  0.723
HHIncomeAgLab     |  0.499 11.686  0.249 | -0.198  2.646  0.039 |  0.057  0.259
NonAgLabInc       |  0.063  0.187  0.004 |  0.086  0.500  0.007 |  0.233  4.264
WildProdInc       |  0.032  0.049  0.001 | -0.105  0.749  0.011 | -0.140  1.541
SheaInc           |  0.564 14.958  0.319 |  0.152  1.554  0.023 |  0.147  1.702
CottonInc         |  0.067  0.212  0.005 |  0.031  0.065  0.001 | -0.062  0.301
AgTradeInc        |  0.532 13.313  0.284 |  0.284  5.450  0.081 | -0.031  0.075
NonAgTradeInc     |  0.141  0.938  0.020 | -0.078  0.413  0.006 |  0.331  8.580
                    cos2  
HHIncomePrimary    0.007 |
HHIncomeFarm       0.001 |
HHIncomeLivestock  0.009 |
HHIncomeAgLab      0.003 |
NonAgLabInc        0.054 |
WildProdInc        0.020 |
SheaInc            0.022 |
CottonInc          0.004 |
AgTradeInc         0.001 |
NonAgTradeInc      0.109 |

7.4 Savelugu (Clustering)

This section depicts cluster analysis of income diversification in Savelugu.

8.0 Comparative Analysis between Kumbungu and Savelugu

The Principal Component Analysis (PCA) results from both Kumbungu and Savelugu provide valuable insights into the socio-economic dynamics of these communities, which are crucial for understanding urban climate challenges and vulnerabilities. Both analyses reveal significant differences and similarities in terms of key socio-economic factors influencing climate adaptation strategies.

8.1 Socio-Economic Dimensions and Variance Explained

In Kumbungu, the first three principal components (PCs) together explain over 35% of the total variance, with the first component contributing 14.6%. In contrast, Savelugu’s first three PCs explain over 41% of the variance, with Dim.1 alone contributing 18.29%. This suggests that Savelugu’s socio-economic data has a broader diversity of influencing factors, as reflected in the higher cumulative variance explained by the first three components.

The variance explained in Kumbungu’s PCA is relatively lower, indicating that there are fewer prominent socio-economic factors in the data that capture significant variability. However, the first component’s focus on variables like “LandOwnership,” “RemitYesCash,” and “RemitYesKind” in Kumbungu suggests a strong relationship between land tenure systems and remittance flows. These factors are closely tied to urban adaptation strategies, as land ownership and remittances directly influence people’s ability to cope with climate risks, such as heatwaves and flooding.

In contrast, Savelugu’s PCA reveals a more nuanced socio-economic landscape with contributions from remittance flows, land access, and household mobility. The significant contributions of “RemitYesCash” and “LandAggrAcc” to Dim.1 in Savelugu indicate that socio-economic mobility and land tenure play a prominent role in shaping urban climate vulnerability. These factors may influence how households in Savelugu respond to environmental changes, emphasizing the need for urban planning that accounts for both economic and land access challenges.

8.2 Income Diversification and Climate Resilience

Income diversification is a critical factor in shaping urban climate resilience, and the PCA results from both Kumbungu and Savelugu highlight this trend. In Kumbungu, the first nine components explain over 63% of the variance in income sources, with significant contributions from agricultural activities like “HHIncomeFarm” and “SheaInc.” The PCA results suggest that households in Kumbungu rely heavily on primary income from farming, which is directly linked to the region’s climate vulnerability. This dependence on agriculture implies that climate adaptation strategies need to focus on enhancing agricultural sustainability, improving land management, and promoting alternative income sources, such as shea production.

On the other hand, Savelugu’s income diversification pattern reveals a broader reliance on both agricultural and non-agricultural income sources. The first two principal components (Dim.1 and Dim.2) in Savelugu show substantial contributions from agricultural income, but non-agricultural income sources like “NonAgTradeInc” and “WildProdInc” also play a significant role in Dim.3. The increasing reliance on non-agricultural income streams in Savelugu suggests that the community is adapting to climate change by diversifying income sources beyond agriculture. This economic shift highlights the need for urban climate policies that support both agricultural sustainability and the growth of non-agricultural sectors, particularly in the face of climate pressures.

8.3 Land Tenure and Adaptation to Urban Climate Risks

In terms of land tenure, Kumbungu’s PCA emphasizes the importance of “LandOwnership” and land tenure systems in influencing urban adaptation strategies. The contribution of land ownership to the first principal component indicates that access to land is a key factor in determining how households in Kumbungu respond to urban climate risks. This finding highlights the role of land policies in fostering resilience, as secure land tenure can empower households to make long-term investments in climate adaptation.

In Savelugu, land access is also a critical factor, with variables like “LandAggrAcc” contributing to Dim.1. However, the relationship between land tenure and climate adaptation is more complex in Savelugu due to the higher diversity of income sources and mobility patterns. This complexity suggests that in Savelugu, urban adaptation strategies must be more multifaceted, addressing both land access and the growing role of non-agricultural income streams.

8.4 Remittances as a Critical Factor in Urban Climate Adaptation

Remittances, both in cash and kind, emerge as a critical variable in both Kumbungu and Savelugu. In Kumbungu, “RemitYesCash” and “RemitYesKind” contribute significantly to the first principal component, suggesting that remittance flows play an important role in supporting households’ economic resilience to climate risks. Similarly, in Savelugu, “RemitYesCash” strongly correlates with Dim.1, indicating the importance of remittances in shaping urban climate dynamics. This suggests that remittances not only provide immediate economic relief but also contribute to long-term climate adaptation strategies by enabling households to invest in climate-resilient infrastructure and livelihoods.

8.5 Policy Implications for Urban Climate Resilience

The PCA analyses for both Kumbungu and Savelugu offer valuable insights into the socio-economic and land tenure factors that shape urban climate resilience. In Kumbungu, where the socio-economic data exhibits lower variance, interventions should focus on improving agricultural sustainability and securing land tenure. In Savelugu, the higher variance in the data suggests that a broader range of socio-economic factors must be addressed, including promoting non-agricultural income sources and improving access to land.

In both communities, remittances emerge as a key factor in enhancing urban resilience to climate change. Urban climate policies should, therefore, recognize the role of remittances in bolstering households’ adaptive capacities. Additionally, promoting economic diversification, enhancing land tenure security, and supporting sustainable agricultural practices will be essential in developing effective climate adaptation strategies in both Kumbungu and Savelugu.

By comparing the PCA results from both communities, it becomes evident that urban climate challenges in Northern Ghana are influenced by a complex interplay of socio-economic, land tenure, and income diversification factors. Understanding these dynamics is crucial for designing context-specific policies that enhance urban resilience and adaptive capacity in the face of climate change.

9.0 Conclusion

In conclusion, the PCA analyses for both Kumbungu and Savelugu reveal distinct socio-economic and land tenure factors influencing urban climate adaptation in these communities. Kumbungu’s analysis emphasizes the role of remittances and land ownership, highlighting their significance in climate resilience strategies. In contrast, Savelugu’s findings show a strong reliance on agricultural income and growing diversification into non-agricultural sources as a response to climate challenges. These insights provide valuable guidance for targeted climate adaptation interventions tailored to the specific socio-economic dynamics of each area.