A Simple Poverty Scorecard for the Philippines

Mark Schreiner

April 16, 2008

Senior Scholar, Center for Social Development

Washington University in Saint Louis
Campus Box 1196, One Brookings Drive
Saint Louis, MO 63130–4899, U.S.A.
and
Director, Microfinance Risk Management, L.L.C.
2441 Tracy Avenue, Kansas City, MO 64108–2935, U.S.A.
Telephone: +1 (816) 359–3545, http://www.microfinance.com

Abstract
How poor are participants of development projects in the Philippines? This paper uses
the 2002 Annual Poverty Indicators Survey to construct an easy-to-use, objective
poverty scorecard that estimates the likelihood that a participant has income below the
national poverty line. The scorecard uses 10 simple indicators that field workers can
quickly collect and verify. Scores can be computed by hand on paper in real time. With
99-percent confidence, estimates of groups’ overall poverty rates are accurate to within
+/–1 percentage points. The poverty scorecard can help programs target services, track
changes in poverty over time, and report on poverty rates.

Version note
This version replaces that of April 27, 2007, which, due to an incorrect Purchase Power
Parity conversion factor, used an incorrect $2/day line. The correct $2/day line turns
out to be almost the same as the “very poor” line defining the poorest half below the
national poverty line. The $3/day line, in turn, is almost exactly the national poverty
line. Therefore, this version reports on the $4/day line. These changes affect Section 8
and Figures 26–29; everything else is unchanged. Organizations that have applied the
poverty scorecard here should simply discard their $2/day figures and associate their
existing scores with the correct poverty likelihoods for $4/day, using Figure 26.

Acknowledgements
This paper was commissioned by Grameen Foundation USA and funded by the
Consultative Group to Assist the Poorest under the CGAP/Ford Social Indicators
Project. Data were provided by the National Statistics Office of the Republic of the
Philippines. Thanks go to Aniceta Alip, Malika Anand, Nigel Biggar, Cecilia del
Castillo, Ron Chua, Melany Grecia-Viajante, Gretel Guzmán, Syed Hashemi, Lalaine
Joyas, Anthony Leegwater, Gilbert Maramba, Lourdes Medina, Vincent Morris D.
Olaivar, Luca David Opramolla, Don Sillers, and Jeff Toohig.
 A Simple Poverty Scorecard for the Philippines

1. Introduction

This paper presents an easy-to-use, objective poverty scorecard to help

development programs in the Philippines to target services, track changes in poverty

over time, and report clients’ poverty rates.

Indicators in the scorecard were derived from the 38,014 households surveyed in

the 2002 Annual Poverty Indicators Survey (APIS). Selection criteria included:

• Inexpensive to collect, easy to answer quickly, and simple to verify

• Liable to change over time as poverty status changes

• Strongly correlated with poverty

All scorecard weights are positive integers, and scores range from 0 (most-likely

“poor”) to 100 (least-likely “poor”). The scorecard is easy to understand, and field

workers can compute scores by hand, on paper, in real time.

A participant’s score corresponds to a “poverty likelihood”, that is, the

probability of being poor. For a group, the overall poverty rate (the so-called “head-

count index”) is the average poverty likelihood of the individuals in the group. For a

group over time, progress (or regress) is the change in its average poverty likelihood.

The poverty scorecard here should qualify for certification for the reporting

required of USAID’s microenterprise partners. In particular, the scorecard is highly

practical to use. Also, it accurately and objectively estimates the likelihood of having

income below the national poverty line. With 90-percent confidence, a household’s
estimated poverty likelihood is accurate within +/–6 percentage points, and a group’s

estimated overall poverty rate is accurate with 99-percent confidence to within +/–1

percentage points.

2. Data and poverty lines

The analysis uses the 38,014 households in the 2002 APIS from the Philippines’

National Statistics Office. This is the best, most recent household survey available with

income or expenditure data.

This paper divides the APIS households into three random samples (Figure 1),

with one-half the households used for constructing the scorecard, one-fourth used for

associating scores with estimated poverty likelihoods, and one-fourth used for measuring

the accuracy of estimates derived from the scorecard.

APIS is fielded annually and measures income but not expenditure. The official

poverty lines are in terms of income, and the Philippine government applies them only

to a larger, more detailed survey, the triennial Family Income and Expenditure Survey

(FIES). The 2003 FIES is not available, but Ericta (2005) reports that it gives a

poverty rate of 30.4 percent.

This paper applies the official poverty lines to the income meaure in the 2002

APIS. While APIS uses different questions than FIES to measure income, the resulting

overall poverty rate is 31.8 percent, remarkably close to FIES’ 30.4 percent.

1
 The rural poverty rate in APIS was 46.4 percent, while urban was 17.3 percent.

This paper presents a single scorecard for use anywhere in the Philippines, as evidence

from India and Mexico (Schreiner, 2006 and 2005a) suggests that there are only small

returns to segmenting scorecards by rural and urban.

Figure 2 shows the official poverty lines by urban/rural for each province. It also

shows the “half lines” that demarcate the very poor, that is, the poorest half of the poor.

The second-to-last section of the paper looks at poverty by the $4/day-or-less standard.

3. Scorecard construction

About 500 potential poverty indicators were prepared, including:

• Household and housing characteristics (such as cooking fuel and type of floor)

• Individual characteristics (such as age and highest grade completed)

• Household consumption (such as spending on non-alcoholic drinks)

• Household durable goods (such as electric fans and telephones)

Each indicator’s ability to predict poverty was tested first with the entropy-

based “uncertainty coefficient” (Goodman and Kruskal, 1979). This resembles a

correlation coefficient, but it is applied to categorical indicators (such as “type of floor”)

rather than continuous ones (such as “square meters of floor space”). About 120

indicators were then selected for further analysis. Figure 3 lists the top 50, ranked by

uncertainty coefficient. Responses are ordered by strength of association with poverty.

2
 Many indicators in Figure 3 are similar in terms of their link with poverty. For

example, most households who have a television also have electricity. If a scorecard

already includes “has a television”, then “has electricity” is superfluous. Thus, many

indicators strongly linked with poverty are not in the scorecard because similar

indicators are already included.

The scorecard also aims to measure changes in poverty through time. Thus, some

powerful indicators (such as education of the female head/spouse) that are unlikely to

change as poverty changes were omitted in favor of slightly less-powerful indicators

(such as the number of radios) that are more likely to change. All the indicators of

consumption (such as “In the past six months, how much on average per week did the

household spend on dairy products and eggs”) were not selected because they cannot be

directly observed nor verified.

The scorecard itself was constructed using Logit regression. Indicator selection

combined statistics with the judgment of an analyst with expertise in scoring and

development. Starting with a scorecard with no indicators, each candidate indicator was

added, one-by-one, to a one-indicator scorecard, using Logit to derive weights. The

improvement in accuracy for each indicator was recorded using the “c” statistic.1

1
Higher “c” indicates greater ability to rank households by poverty status. For a Logit
regression with a categorical outcome (such as poor/not poor), “c” is a general measure
of explanatory power, much like R2 in a least-squares regression on a continuous
outcome. “c” is equal to the Mann-Whitney statistic (also known as the Wilcoxon rank-
sum statistic) that indicates how much two distributions overlap (here, the distributions
are of the estimated poverty likelihoods for poor and non-poor households). “c” is also
equivalent to the area under an ROC curve—discussed in more detail later—that plots

3
 After all indicators had been tested, one was selected based on several factors

(Schreiner et al., 2004; Zeller, 2004). These included the improvement in accuracy, the

likelihood of acceptance by users (determined by simplicity, cost of collection, and “face

validity” in terms of experience, theory, and common sense), the ability of the indicator

to change values as poverty status changes over time, variety vis-à-vis other indicators

already in the scorecard, and ease of observation/verification.

The selected indicator was then added to the scorecard, and the previous steps

were repeated until 10 indicators were selected. Finally, the Logit coefficients were

transformed into non-negative integers such that the lowest possible score is 0 (most

likely poor) and the highest is 100. The final poverty scorecard appears in Figure 4.

This statistical algorithm is the Logit analogue to the stepwise “MAXR” in, for

example, Zeller, Alcaraz and Johannsen (2005) and IRIS (2005a and 2005b). The

procedure here diverges from naïve stepwise in that expert judgment and non-statistical

criteria were used to select from among the most-predictive indicators. This improves

robustness and, more importantly, helps ensure that the indicators are simple and

sensible, increasing the likelihood of acceptance by users.

the share of poor and non-poor households versus all households ranked by score.
Finally, “c” can also be seen as the share of all possible pairs of poor and non-poor
households in which the poor household has a lower score. The more often the poor
household has the lower score, the better the ranking by poverty status.

4
4. Scorecard use

As explained in Schreiner (2005b), the central challenge is not to maximize

accuracy but rather to maximize the likelihood of programs’ using scoring

appropriately. When scoring projects fail, the culprit is usually not inaccuracy but

rather the failure of users to accept scoring and to use it properly (Schreiner, 2002). The

challenge is not technical but human and organizational, not statistics but change

management. “Accuracy” is easier—and less important—than “practicality”.

The scorecard here was designed to help users to understand and trust it (and

thus use it properly). While accuracy matters, it must be balanced against simplicity,

ease-of-use, and “face validity”. In particular, programs are more likely to collect data,

compute scores, and pay attention to the results if, in their view, scoring avoids creating

“extra” work and if the whole process generally seems to make sense.

This practical focus naturally leads to a one-page scorecard (Figure 4) that

allows field workers to score households by hand in real time because it features:

• Only 10 indicators

• Only observable, categorical indicators (“flooring material”, not “value of house”)

• User-friendly weights (non-negative integers, no arithmetic beyond simple addition)

Among other things, this simplicity enables “rapid targeting”, such as

determining (in a day) who in a village qualifies for, say, work-for-food, or ration cards.

5
 The scorecard in Figure 4 can be photocopied for immediate use. It can also

serve as a template for data-entry screens with database software that records

indicators, indicator values, scores, and poverty likelihoods.

A field agent collecting data and computing scores on paper would:

• Read each question off the scorecard

• Circle the response and the corresponding points

• Write the points in the far-right column

• Add up the points to get the total score

• Implement program policy based on the score

4.1 Scores and poverty likelihoods

A score is not a poverty likelihood (that is, the probability of being poor), but

each score is associated with an estimated poverty likelihood via a simple table (Figure

5). For example, scores of 25–29 correspond to a poverty likelihood of 76.8 percent.

6
 Scores (sums of weights) are associated with estimated poverty likelihoods

(probabilities of being poor) via the “bootstrap” (Efron and Tibshirani, 1993):

• From the first one-fourth hold-out sample, draw a new sample of the same size with

replacement

• For people in a given score range, compute the share who are poor

• Repeat the previous two steps 10,000 times

• For a given score range, define the poverty likelihood as the average of the shares of

people who are poor in that score range across the 10,000 samples

These resulting poverty likelihoods are objective, that is, based on data. This

process would produce objective poverty likelihoods even if the scorecards themselves

were constructed without data. In fact, scorecards of objective, proven accuracy are

often constructed only with qualitative judgment (Fuller, 2006; Caire, 2004; Schreiner et

al., 2004). Of course, the scorecard here uses data. While its construction—like any

statistical analysis—was partially informed by the analyst’s judgment, the explicit

acknowledgment of this fact is irrelevant for the objectivity of the poverty likelihoods.

After all, objectively depends on using data to associate scores with poverty likelihoods,

not on pretending to avoid the use of judgment during scorecard construction.

Figure 6 depicts the precision of estimated poverty likelihoods as point estimates

with 90-, 95-, and 99-percent confidence intervals. This is the standard way to measure

accuracy, and it is widely understood by lay people. The confidence intervals here were

derived empirically from the 10,000 bootstrap samples described above. For a given

7
score, the lower (upper) bound on the x-percent confidence interval is the value less

(greater) than (100–x)/2 percent ((100+x)/2 percent) of the bootstrapped likelihoods.

For example, the average poverty rate across bootstrap samples for people with

scores of 25–29 is 76.8 percent (this is the poverty likelihood in Figure 5). In 90 percent

of samples, the poverty rate is between 73.1–80.4 percent (Figure 6). In 95 percent of

samples, the share is 72.4–81.0; in 99 percent of samples, the share is 70.8–82.3.

For estimated and true poverty likelihoods, Figure 7 depicts mean absolute

differences and confidence intervals from bootstrapping the second one-fourth hold-out

sample from the 2002 APIS. The mean absolute difference is 3.6 percentage points.

This discussion so far looks at whether estimated poverty likelihoods are close to

true poverty likelihoods (and indeed they are). There is another aspect of accuracy, one

associated with targeting: how well the poor are concentrated in low scores. A perfect

scorecard would assign all the lowest scores to poor people (and all the highest scores to

non-poor people). In reality, no scorecard is perfect, so some poor people have high

scores, and vice versa.

ROC curves are standard tools for showing how well the poor are concentrated

in lower scores (Baulch, 2003; Wodon, 1997). They plot the share of poor and non-poor

households against the share of all households ranked by score.

What does the ROC curve in Figure 8 mean? Suppose a program sets a cut-off

so as to target the lowest-scoring x percent of people. The ROC curve then shows the

share of the poor (northwest curve) and non-poor (southwest curve) targeted. Greater

8
ability to rank-order—with less leakage and less undercoverage—is shown by curves

that are closer to the northwest and southeast corners of the graph.

In Figure 8, the northwest (southeast) curve depicts accuracy among the poor

(non-poor). As a benchmark, the external trapezoid shows the accuracy of a

hypothetical perfect scorecard that assigns all of the lowest scores to poor people. The

diagonal line represents random targeting.

The curves for the scorecard show, for example, that targeting the 20 percent of

households with the lowest scores would target 51 percent of all the poor and 6 percent

of all the non-poor. In contrast, randomly targeting 20 percent of cases would target 20

percent of the poor and 20 percent of the non-poor.

Figure 8 also reports two other common measures of rank-ordering. The first is

the Kolmogorov-Smirnov (KS) statistic, defined as the maximum distance between the

poor and non-poor curves (here 59.2). Higher KS implies better rank-ordering.

The second measure is the ratio of the area inside the ROC curves to the area

inside the trapezoid of a hypothetical perfect scorecard (here 75.5). Again, greater area

within the curves implies better rank-ordering.

Is this scorecard accurate enough for targeting? Errors due to scorecard

inaccuracy are probably small relative to errors due to other sources (such as mistakes

in data collection or fraud) and relative to the accuracy of other feasible targeting tools.

All in all, Figures 6–8 suggest that the estimated likelihoods of being poor are estimated

both accurately and precisely.

9
4.2 Estimates of overall poverty rates

The estimated overall poverty rate is the average of the estimated poverty

likelihoods of individuals.

For example, suppose a program has 3,000 participants on Jan. 1, 2006 and that

1,000 have scores of 20, 1,000 have scores of 30, and 1,000 have scores of 40. The

poverty likelihoods that correspond to these scores are 77.6, 77.7 and 48.3 percent

(Figure 5). The overall poverty rate is the participants’ average poverty likelihood, that

is, 1,000 x (77.6 + 77.7 + 48.3) ÷ 3,000 = 67.9 percent.

To test accuracy and precision, the scorecard was applied to 10,000 bootstrap

replicates from the second one-fourth hold-out sample, comparing the estimated overall

poverty rates with the true values. The mean difference was 0.1 percentage points, with

a standard deviation of 0.37. The 90-percent confidence interval around the mean was

+/–0.6 percentage points, the 95-percent interval was +/–0.7 percentage points, and

the 99-percent interval was +/–1.0 percentage points. The estimated overall poverty

rate is thus unbiased and highly precise.

4.3 Progress out of poverty over time

For a given group, progress out of poverty over time is estimated as the change

in the average poverty likelihood.

Continuing the previous example, suppose that on Jan. 1, 2007, the same 3,000

people (some of whom may no longer be participants) are now in groups of 500 with

10
scores of 20, 25, 30, 35, 40, and 45 (by Figure 5, poverty likelihoods of 77.6, 76.8, 77.7,

48.6, 48.3, and 33.6 percent). Their average poverty likelihood is now 60.4 percent, an

improvement of 67.9 – 60.4 = 7.5 percentage points. In other words, 7.5 of every 100 in

this group left poverty. Among those who were poor to start with, one in nine (7.5 ÷

67.9 = 11.1 percent) left poverty.

Of course, the scorecard does not indicate what caused progress; it just measures

the change, regardless of cause.

5. Setting targeting cut-offs

How would the poverty scorecard be used for targeting? Potential participants

with scores at or below a targeting cut-off are labeled targeted and treated—for

program purposes—as if they were poor. Those with higher scores are non-targeted and

treated—again, for program purposes—as if they were non-poor.

Poverty status (expenditure below a poverty line) is distinct from targeting status

(score below a cut-off). Poverty status is a fact whose determination requires an

expensive survey. In contrast, targeting status is a policy choice whose determination

requires a cut-off and an inexpensive estimate of poverty likelihood. Indeed, the purpose

of scoring is to infer poverty status without incurring the cost of direct measurement.

No scorecard is perfect, so some of the truly poor will not be targeted, and some

of the truly non-poor will be targeted. Targeting is accurate to the extent that poverty

status matches targeting status. In turn, this depends on the selection of a targeting

11
cut-offs and how it balances accuracy for the poor versus non-poor. The standard

approach uses a classification matrix and a net-benefit matrix (SPSS, 2003; Adams and

Hand, 2000; Salford Systems, 2000; Hoadley and Oliver, 1998; Greene, 1993).

5.1 Classification matrix

Given a targeting cut-off, there are four possible classification results:

A. Truly poor correctly targeted (score at or below the cut-off)

B. Truly poor mistakenly non-targeted (score above cut-off)

C. Truly non-poor mistakenly targeted (score at or below cut-off)

D. Truly non-poor correctly non-targeted (score above cut-off)

These four possibilities can be shown as a general classification matrix (Figure

9). Accuracy improves as there are more cases in A and D and fewer in B and C.

Figure 10 shows the number of people in each classification by score in the

second one-fourth hold-out sample. For example, with a cut-off of 25–29, there are:

A. 12.4 truly poor correctly targeted

B. 19.7 truly poor mistakenly non-targeted

C. 2.2 truly non-poor mistakenly targeted

D. 65.7 truly non-poor correctly non-targeted

12
 Targeting accuracy (and errors of undercoverage and leakage) depends on the

cut-off. For example, if the cut-off were increased to 39–34, more poor (but less non-

poor) are correctly targeted:

A. 16.6 truly poor correctly targeted

B. 15.5 truly poor mistakenly non-targeted

C. 3.4 truly non-poor mistakenly targeted

D. 64.5 truly non-poor correctly non-targeted

Whether a cut-off of 40–44 is preferred to 45–49 depends on net benefit.

5.2 Net-benefit matrix

Each of the four classification results is associated with a net benefit (Figure 11):

α. Benefit per truly poor person correctly targeted

β. Cost (negative net benefit) per truly poor person mistakenly non-targeted

γ. Cost (negative net benefit) per truly non-poor person mistakenly targeted

δ. Benefit per truly non-poor person correctly non-targeted

Each net benefit α, β, γ, and δ corresponds to one of the quadrants in the general

classification matrix in Figure 9. For example, α is the net benefit associated with each

truly poor person who is correctly targeted (quadrant A), and β is the cost (negative net

benefit) associated with each truly poor person incorrectly targeted (quadrant B).

13
 Given a net-benefit matrix and a classification matrix, total net benefit is the

sum of the net benefit per person in each quadrant multiplied by the number of people

in the quadrant, summed across all four quadrants:

Total net benefit = α·A + β·B + γ·C + δ·D.

To set an optimal cut-off, a program would:

• Select a net-benefit matrix based on its values and mission

• Compute total net benefits for each cut-off with the net-benefit matrix and Figure 10

• Select the cut-off with the highest total net benefit

The only non-trivial step is selecting a net-benefit matrix. Some common net-

benefit matrices are discussed below. In general, however, each program should

thoughtfully decide for itself how much it values successful targeting versus errors of

undercoverage and leakage. Of course, any program that targets already uses (if only

implicitly) a net-benefit matrix. It is healthy to go through a process of thinking

explicitly and intentionally about the value of possible targeting outcomes.

For example, suppose a program places great importance on correctly targeting

the poor, even at the cost of accidentally targeting more non-poor. It could reflect this

valuation by increasing the weight on quadrant A (by increasing its net benefit α),

and/or by decreasing the weight on quadrant B (by decreasing its net benefit β). The

examples of net-benefit matrices discussed next represent different valuations of

correctly/incorrectly targeting the poor/non-poor.

14
5.2.1 “Total Accuracy”

As an example, suppose a program selects the net-benefit matrix that

corresponds to the “Total Accuracy” criterion (Figure 12, IRIS, 2005b). Then total net

benefit is the number of people correctly classified:

Total net benefit = 1·A + 0·B + 0·C + 1·D,

= A + D.

This values correct classifications of the poor and non-poor equally. Grootaert

and Braithwaite (1998) and Zeller, Alcaraz, y Johannsen (2005) use “Total Accuracy”

to evaluate their poverty scorecards.

Figure 13 shows “Total Accuracy” for all cut-offs. Total net benefit is greatest

(81.1) for a cut-off of 30–34; at that point, poverty segment matches poverty status for

four out of five people.

“Total Accuracy” weighs the poor and non-poor the same. If most people are

non-poor and/or if a scorecard is more accurate for the non-poor, then “Total Accuracy”

might look good even if few poor people are correctly classified. Development programs,

however, probably value correct targeting more for the poor than for the non-poor.

A simple, transparent way to reflect this valuation is to increase the relative net

benefit α of correctly classifying the poor. For example, if a program values correctly

targeting the poor twice as much as correctly not targeting the non-poor, then α should

be set twice as high as δ in the net-benefit matrix. Then the new optimal cut-off is 50–

54, the cut-off point where α.A + δ.D = 2.A + D is highest.

15
5.2.2 “Poverty Accuracy”

A criterion that values only correctly classifying the poor is “Poverty Accuracy”

(Figure 14, IRIS, 2005b):

Total net benefit = 1·A + 0·B + 0·C + 0·D,

= A.

Of course, correctly targeting the poor is rarely the sole criteria. In fact, Figure

13 shows that “Poverty Accuracy” is greatest with a cut-off of 95–100. While targeting

everyone does ensure that all poor people are targeted and so minimizes undercoverage

of the poor (second-to-last column of Figure 13), it also targets all the non-poor and so

maximizes leakage (the last column of Figure 13).

5.2.3 “Non-poverty Accuracy”

“Non-poverty Accuracy” counts only correct classifications of the non-poor (total

net benefit is D). This is maximized by setting a cut-off of 0–4 and thus not targeting

anyone (minimum leakage but maximum undercoverage).

5.2.4 “BPAC”

IRIS (2005b) proposes a new measure of accuracy called the “Balanced Poverty

Accuracy Criterion”. BPAC balances two goals:

• Accuracy of the estimated overall poverty rate

• “Poverty Accuracy”

According to IRIS (2005b), the first goal is optimized when undercoverage B is

balanced by leakage C, and the second goal is optimized by maximizing A. If B > C,

16
then BPAC’s net-benefit matrix is Figure 15. In essence, BPAC maximizes A while

making B and C as close to each other as possible:

Total net benefit = 1·A + 1·B + (–1)·C + 0·D,

= A + (B – C).

If C > B, then total net benefit under BPAC is A + (C – B).

BPAC was invented because IRIS does not estimate poverty likelihoods. Instead,

IRIS estimates expenditure and then labels as poor those households with estimated

expenditure less than the poverty line. In this set-up, the overall poverty rate is

estimated as the share of people targeted, and this estimate is most accurate (that is,

closest to the true value) when undercoverage B equals leakage C.

For a scorecard (like the one here) that estimates poverty likelihoods, however,

BPAC is not meaningful. This is because the estimated overall poverty rate is the

average of participants’ estimated poverty likelihoods. These estimates are independent

of whatever targeting cut-off a program might set. In contrast, the targeting errors of

undercoverage B and leakage C depend directly on the cut-off. Thus, for scorecards that

estimate poverty likelihoods, getting B close to C is not related to optimizing the

accuracy of the estimated overall poverty rate and so is not related to BPAC’s goals.

17
6. Training, quality-control, and MIS

The technical aspects of scorecard construction and accuracy just discussed are

important, but gaining the trust and acceptance of managers and field workers is even

more important (Schreiner, 2002).

In particular, the field workers who collect indicators must be trained. If they put

garbage in, the scorecard will put garbage out. To prevent abuse, on-going quality

control of data is required.

Programs should record in their MIS at least the poverty likelihood along with

an identifier for each client. Ideally, they would also record the score, the indicators,

and the values of the indicators. This will allow quick computation of average poverty

likelihoods (as well as other analyses), both for a point in time and for changes through

time (Matul and Kline, 2003).

7. Calibrating the scorecard for the very poor

The simple poverty scorecard in Figure 4 can be used to track outreach not only

to the poor but also to the very poor, that is, the poorest half of the poor below the

national poverty line. This is the relevant group for USAID certification.

18
7.1 Poverty likelihoods

As before, scores are associated with the probability of being very poor by

bootstrapping 10,000 samples from first one-fourth hold-out sample from the 2002

APIS. The poverty likelihood for a given score is then taken as the average of the

shares of people with that score who are very poor across the 10,000 samples.

Columns 2–4 in Figure 16 are the poverty likelihoods for the three classes for all

scores. For example, if a potential participant has a score of 25–29, the probability of

being very poor is 42.2 percent, the probability of being poor is 34.4 percent, and the

probability of being non-poor is 23.2 percent.

Columns 5–7 in Figure 16 are the share of targeted participants by poverty

status and by cut-off. For example, for a cut-off of 25–92, 58.3 percent of those targeted

would be very poor, 26.4 percent would be poor, and 15.2 percent would be non-poor.

Each person is associated with three poverty likelihoods. For example, a person

with a score of 25 may be targeted as very poor, but the likelihood of truly being very

poor is not 100 percent but rather 42.4 percent (from Figure 16). The same person has a

34.4-percent likelihood of being truly poor, and a 23.2-percent likelihood of being truly

non-poor. Each person has one targeting status (for program purposes), one true

poverty status (in reality), and three estimated poverty likelihoods (one for each

possible poverty status).

As before, these poverty likelihoods are objective, that is, based on data. They

are valid even though the scorecard was not constructed originally to predict the

19
likelihood of being very poor. It works because the likelihood of being very poor is

highly correlated with having a low score (high likelihood of being poor). A scorecard

could be built specifically for the very poor, but it would add cost and complexity.

Figure 17 shows the precision of estimated poverty likelihoods for being very

poor as point estimates with 90-, 95-, and 99-percent confidence intervals. For example,

the average poverty rate (the poverty likelihood) across bootstrap samples for people

with scores of 25–29 was 42.4 percent. In 90 percent of 10,000 bootstraps from the

second one-fourth hold-out sample, the share was between 38.2–46.7 percent. In 95

percent of samples, the share was between 37.3–47.7, and in 99 percent of samples, the

share was between 35.9–49.3.

For estimated and true poverty likelihoods, Figure 18 depicts mean absolute

differences and confidence intervals from 10,000 bootstraps on the second one-fourth

hold-out sample. Weighting by the people in a score range, the mean absolute difference

is 1.8 percentage points, with a 90-percent interval of +/–2.5 percentage points.

The other aspect of accuracy is how well the very poor are concentrated in low

scores. Once again, an ROC curve is a useful way to look at this.

Figure 19 plots the share of the very poor against the share of the not very poor,

ranked by score. For example, targeting the 30 percent of cases with the lowest scores

would target 77 percent of all the very poor and 21 percent of all the not very poor.

20
 In terms of the Kolmogorov-Smirnov statistic, the maximum distance between

the curves is 56.0. In terms of the ratio of the area inside the scorecard curves to the

area inside the trapezoid of a hypothetical perfect scorecard, the value is 73.4.

All in all, Figures 17–19 suggest that the likelihoods of being very poor are

estimated both accurately and precisely.

7.2 Overall poverty rates for the very poor

The average of estimated poverty likelihoods for a group is their estimated

overall (very poor) poverty rate. To measure the accuracy and precision of this

estimate, the scorecard was applied to 10,000 bootstrap replicates from the second one-

fourth hold-out samples from the 2002 APIS, and then the estimated overall poverty

rates were compared with the true values. The mean difference was 0.8 percentage

points, with a standard deviation of 0.32. The 90-percent confidence interval around the

mean was +/–0.5 percentage points, the 95-percent interval was +/–0.6 percentage

points, and the 99-percent interval was +/–0.8 percentage points.

Thus, subtracting 0.8 percentage points to a group’s average poverty likelihood

would produce an unbiased estimate that, in 99 of 100 cases, would be within +/–0.8

percentage points of the true overall (very poor) poverty rate. This estimate is both

quite accurate and quite precise.

21
7.3 Targeting the very poor

As before, targeting involves using a classification matrix and a net-benefit

matrix to select a cut-off. The wrinkle is that there are now three poverty statuses:

• Very poor: Poorest half of those with expenditure at or below the poverty line

• Poor: Least-poor half of those with expenditure at or below poverty

• Non-poor: Expenditure above poverty

There are also three targeting segments:

• Very poor: Score at or below the very poor/poor cut-off

• Poor: Score above the very poor/poor cut-off and

at or below the poor/non-poor cut-off

• Non-poor: Score above the poor/non-poor cut-off

22
 There are two cut-offs (very poor/poor and poor/non-poor) and 9 classification

results (Figure 20):

A. Truly very poor correctly targeted as very poor

B. Truly very poor incorrectly targeted as poor

C. Truly very poor incorrectly targeted as non-poor

D. Truly poor incorrectly targeted as very poor

E. Truly poor correctly targeted as poor

F. Truly poor incorrectly targeted as non-poor

G. Truly non-poor incorrectly targeted as very poor

H. Truly non-poor incorrectly targeted as poor

I. Truly non-poor correctly targeted as non-poor

The general classification matrix (Figure 20) and the net-benefit matrix (Figure

21) are combined as before to define total net benefit:

Total net benefit = α·A + β·B + γ·C + δ·D + ε·E + ζ·F + η·G + θ·H + ι·I.

Figure 22 shows classification results for all possible pairs of cut-off scores in the

second one-fourth hold-out sample. For example, suppose a program defined:

• Very poor/poor cut-off of 20–24 (so scores of 0–24 are targeted as very poor)

• Poor/non-poor cut-off of 30–34 (so scores of 25–34 are targeted as poor, and scores

of 35–100 are targeted as non-poor)

23
 As with any scorecard and cut-offs, there is both successful targeting and errors.

For the example cut-offs of 20–24 and 30–34, targeting would be correct for 65 percent

of the very poor, 39 percent of the poor, and 81 percent of the non-poor (Figure 23).

The program chooses a set of cut-offs to optimize the benefits of correct

classifications, net of the costs (negative benefits) of incorrect classifications. For

example, suppose the net-benefit matrix is Figure 24, representing one way to reflect:

• Greater importance of correctly targeting the very poor and poor

• Greater cost of gross errors such as targeting the truly very poor as non-poor

Given the classification results in Figure 23 and net benefits in Figure 24, total

net benefit for the cut-off pair of 20–24 and 30–34 is +404 (Figure 25).

Is this the best pair of cut-offs? The answer requires applying the net-benefit

matrix to the classification results for all 190 possible pairs (Figure 22). It turns out

that total net benefit is highest for cut-offs 30–34 and 50–54, giving a net benefit of 764.

8. Calibrating for $4/day-or-less poverty

The simple poverty scorecard in Figure 4 can be used to track outreach not only

to the poor (the upper half of those under the national poverty line) and the very poor

(the lower half of those under the national poverty line) but also the $4PPP/day poor,

that is, those with incomes above the national poverty line but below the

$4/day/person international benchmark at purchase-power parity. The Appendix

24
describes the derivation of a $4PPP poverty line that accounts for differences in cost-of-

living across Filipino provinces and across rural and urban areas.

8.1 Poverty likelihoods

Scores are associated with the probability of being very poor by bootstrapping

1,000 samples from the first one-fourth hold-out sample from the 2002 APIS. The

poverty likelihood for a given score is then taken as the average of the shares of people

with that score who are $4/day-or-less poor across the 1,000 bootstrapped samples.

Columns 2–5 in Figure 26 are the poverty likelihoods for the four classes for all

scores. For example, if a potential participant has a score of 25–29, the probability of

being very poor is 42.2 percent, the probability of being poor is 34.4 percent, the

probability of being $4/day poor is 14.4 percent, and the probability of being non-poor

is 8.8 percent. The sum of the four poverty likelihoods is, of course, 100 percent.

Columns 6–9 in Figure 26 are the share of targeted participants by poverty

status and by cut-off. For example, for a cut-off of 25–29, 58.3 percent of those targeted

would be very poor, 26.4 percent would be poor, 9.5 percent would be $4/day poor, and

5.8 percent would be non-poor.

Each person’s score is associated with four poverty likelihoods. For example, a

person with a score of 25 may be targeted as very poor, but the likelihood of truly being

very poor in this case is not 100 percent but rather 42.4 percent (from Figure 26). The

same person has a 34.4-percent likelihood of being truly poor, a 14.4-percent likelihood

of being $4/day poor, and an 8.8-percent likelihood of being truly non-poor. Each

25
person has one targeting status (for program purposes), one true poverty status (in

reality), and four estimated poverty likelihoods (one for each possible poverty status).

Figure 27 shows the precision of estimated poverty likelihoods for being $4/day-

or-less poor as point estimates with 90-, 95-, and 99-percent confidence intervals. For

example, the average $4/day-or-less poverty rate (the poverty likelihood) across

bootstrap samples for people with scores of 25–29 was 91.2 percent. In 90 percent of

1,000 bootstraps from the second one-fourth hold-out sample, the share was between

88.7–93.6 percent. In 95 percent of samples, the share was between 88.1–94.0, and in 99

percent of samples, the share was between 86.9–94.4.

For estimated and true poverty likelihoods, Figure 28 depicts mean absolute

differences and confidence intervals from 1,000 bootstraps on the second one-fourth

hold-out sample. Weighting by the people in a score range, the mean absolute difference

is 4.7 percentage points, with a 90-percent interval of +/–5.2 percentage points.

The other aspect of accuracy is how well the $4/day-or-less poor are

concentrated in low scores. Once again, an ROC curve is a useful way to look at this.

Figure 29 plots the share of the $4/day-or-less poor against the share of the non-

poor, ranked by score. For example, targeting the 20 percent of cases with the lowest

scores would target 44 percent of all the $4/day-or-less poor and 4 percent of all the

non-poor.

26
 In terms of the Kolmogorov-Smirnov statistic, the maximum distance between

the curves is 58.0. In terms of the ratio of the area inside the scorecard curves to the

area inside the trapezoid of a hypothetical perfect scorecard, the value is 74.3.

8.2 Overall poverty rates for the $4/day-or-less poor

The average of estimated poverty likelihoods for a group is their estimated

overall ($4/day-or-less) poverty rate. To measure the accuracy and precision of this

estimate, the scorecard was applied to 1,000 bootstrap replicates from the second one-

fourth hold-out samples from the 2002 APIS, and then the estimated overall poverty

rates were compared with the true values. The mean difference was 0.6 percentage

points, with a standard deviation of 0.39. The 90-percent confidence interval around the

mean was +/–0.6 percentage points, the 95-percent interval was +/–0.7 percentage

points, and the 99-percent interval was +/–1.0 percentage points.

Thus, subtracting 0.6 percentage points to a group’s average poverty likelihood

would produce an unbiased estimate that, in 99 of 100 cases, would be within +/–1.0

percentage points of the true overall ($4/day-or-less) poverty rate. This estimate is both

quite accurate and quite precise.

27
9. Conclusion

One in three Filipinos is poor. An easy-to-use, inexpensive tool for identifying the

poor could improve targeting and speed progress out of poverty. This paper presents a

simple scorecard that estimates the likelihood that a person is poor.

The scorecard is built and tested using data on 38,014 households from the 2002

APIS. The scorecard is calibrated to estimate the likelihood of being poor (income

below the official line), very poor (poorest half of the poor), $4/day poor (income above

the official line but below the $4/day international benchmark), or non-poor.

Out-of-sample bootstrap tests show that the estimates are both accurate and

precise. For individual poverty likelihoods (whether poor or very poor), estimates are

within 6 percentage points of the true value with 90-percent confidence. For a group’s

overall poverty rate (again, whether poor or very poor), estimates are within 1

percentage points of the true value with 99-percent confidence.

For targeting, programs can use the classification results reported here to select

the best cut-off for their particular values and mission.

Accuracy is important, but ease-of-use is even more important; a perfectly

accurate scorecard is worthless if programs feel daunted by its complexity and so never

even try to use it. For this reason, the scorecard here is kept simple, using 10 indicators

that are inexpensive to collect and that are straightforward to observe and verify.

Indicator weights are either zeros or positive integers, and scores range from 0 (most

likely poor) to 100 (least likely poor). Scores are related to poverty likelihoods via a

28
simple look-up table, and targeting cut-offs are also simple to apply. Thus, users can

not only understand the scorecard, but they can also use it to compute scores in the

field, by hand, in real time.

Overall, the poverty scorecard can help development programs to target services

to the poor, track participants’ progress out of poverty through time, and report on

participants’ overall poverty rate.

29
 Appendix:
Adjusting the $4PPP Poverty Line for Cost-of-Living Differences

$4PPP poverty lines were constructed using the following criteria:

• Account for differences in cost-of-living by province and by rural/urban within each

province
• Match the average of rural and urban $4PPP lines to the all-Philippines $4PPP line
• Match the ratio of rural to urban $4PPP lines to that same ratio for the official
national lines

Basic inputs to the calculation include:

• $4PPP/person/day for all-Philippines in 2002 is 46.85 pesos/person/day

• In 2002, 50.16 percent of the population was rural
• The population-weighted official poverty line in 2002 was 38.49 pesos/person/day
for urban areas and 31.59 pesos/person/day for rural areas

The population-weighted average of rural and urban $4PPP lines should match the all-
Philippines $4PPP line:

46.85 = (0.5016 x Rural $4PPP line) + (0.4984 x Urban $4PPP line).

Furthermore, the ratio of the two lines should match the ratio of the official lines:

(Rural $4PPP line ÷ Urban $4PPP line) = 31.59 ÷ 38.49.

Solving the algebra gives:

• Rural $4PPP line of 42.25 pesos/person/day

• Urban $4PPP line of 51.48 pesos/person/day

To account for cost-of-living across provinces, the official lines for 2002 are then
adjusted by their ratio with the rural or urban $4PPP line. For both rural and urban
areas, the adjustment factor is 1.337. That is, the $4PPP poverty line is 33.7 percent
higher than the national poverty line.

$2/day is about equal to the “very poor” line that defines the poorest half below the
national poverty line, and $3/day is almost exactly the same as the national line.

30
References
Adams, N.M.; and D.J. Hand. (2000) “Improving the Practice of Classifier Performance
Assessment”, Neural Computation, Vol. 12, pp. 305–311.

Baulch, Bob. (2003) “Poverty Monitoring and Targeting Using ROC Curves: Examples
from Vietnam”, IDS Working Paper No. 161,
http://www.ids.ac.uk/ids/bookshop/wp/wp161.pdf.

Caire, Dean. (2004) “Building Credit Scorecards for Small Business Lending in
Developing Markets”, Bannock Consulting,
http://www.microfinance.com/English/Papers/Scoring_SMEs_Hybrid.pdf.

Efron, Bradley; and Robert J. Tibshirani. (1993) An Introduction to the Bootstrap, New
York: Chapman and Hall, ISBN 0–412–04231–2.

Ericta, Carmelita N. (2005) “2004 Annual Poverty Indicators Survey (APIS):

Preliminary Results”,
http://www.census.gov.ph/data/pressrelease/2005/ap2004ptx.html.

Fuller, Rob. (2006) “Measuring Poverty of Microfinance Clients in Haiti”,

http://www.microfinance.com/English/Papers/Scoring_Poverty_Haiti_Ful
ler.pdf.

Goodman, L.A. and Kruskal, W.H. (1979) Measures of Association for Cross
Classification, New York, NY: Springer-Verlag, ISBN 0–38–790443–3.

Greene, William H. (1993) Econometric Analysis: Second Edition, New York, NY:
MacMillan, ISBN 0–02–346391–0.

Grootaert, Christiaan; and Jeanine Braithwaite. (1998) “Poverty Correlates and

Indicator-Based Targeting in Eastern Europe and the Former Soviet Union”,
World Bank Policy Research Working Paper No. 1942, Washington, D.C.,
http://www.worldbank.org/html/dec/Publications/Workpapers/WPS1900ser
ies/wps1942/wps1942.pdf.

Hoadley, Bruce; and Robert M. Oliver. (1998) “Business measures of scorecard benefit”,
IMA Journal of Mathematics Applied in Business and Industry, Vol. 9, pp. 55–
64.

IRIS Center. (2005a) “Accuracy Results for 12 Poverty Assessment Tool Countries”,
http://www.povertytools.org/documents/Accuracy%20Results%20for%2012%
20Countries.pdf.

31
IRIS Center. (2005b) “Notes on Assessment and Improvement of Tool Accuracy”,
http://www.povertytools.org/documents/Assessing%20and%20Improving%20
Accuracy.pdf.

Matul, Michal; and Sean Kline. (2003) “Scoring Change: Prizma’s Approach to
Assessing Poverty”, MFC Spotlight Note No. 4, Warsaw, Poland: Microfinance
Centre for Central and Eastern Europe and the New Independent States,
http://www.mfc.org.pl/doc/Research/ImpAct/SN/MFC_SN04_eng.pdf.

National Statistic Coordination Board. (2004) “Poverty Threshold: P11,906 in 2002”,

http://www.nscb.gov.ph/headlines/2004/30Jan04_povtresh.asp, accessed
Sept. 3, 2005.

Salford Systems. (2000) CART for Windows User’s Guide, San Diego, CA.

Schreiner, Mark. (2006) “Is One Simple Poverty Scorecard Enough for India?”, memo
for Grameen Foundation U.S.A.,
http://www.microfinance.com/English/Papers/Scoring_Poverty_India_Seg
ments.pdf.

Schreiner, Mark (2005a) “Un Indice de Pobreza para México”, memo for Grameen
Foundation U.S.A.,
http://www.microfinance.com/Castellano/Documentos/Scoring_Pobreza_Me
xico.pdf.

Schreiner, Mark. (2005b) “IRIS questions on poverty scorecards”, memo for Grameen
Foundation U.S.A.,
http://www.microfinance.com/English/Papers/Scoring_Poverty_Response_
to_IRIS.pdf.

Schreiner, Mark. (2002) Scoring: The Next Breakthrough in Microfinance? Occasional

Paper No. 7, Consultative Group to Assist the Poorest, Washington, D.C.,
http://www.cgap.org/docs/OccasionalPaper_07.pdf,

Schreiner, Mark; Matul, Michal; Pawlak, Ewa; and Sean Kline. (2004) “Poverty
Scorecards: Lessons from a Microlender in Bosnia-Herzegovina”, Microfinance
Risk Management,
http://www.microfinance.com/English/Papers/Scoring_Poverty_in_BiH_Sh
ort.pdf.

SPSS, Inc. (2003) Clementine 8.0 User’s Guide, Chicago, IL, ISBN 1–56827–333–9.

Wodon, Quentin T. (1997) “Targeting the Poor Using ROC Curves”, World
Development, Vol. 25, No. 12, pp. 2083–2092.

32
Zeller, Manfred. (2004) “Review of Poverty Assessment Tools”, Accelerated
Microenterprise Advancement Project,
http://www.povertytools.org/documents/Review%20of%20Poverty%20Assess
ment%20Tools.pdf.

Zeller, Manfred; Alcaraz V., Gabriela; and Julia Johannsen. (2005) “Developing and
Testing Poverty-Assessment Tools: Results from Accuracy Tests in Peru”,
Accelerated Microenterprise Advancement Project,
http://www.povertytools.org/documents/Peru%20Accuracy%20Report.pdf.

33
Figure 1: Households surveyed, people represented,
and overall poverty rates
Sub-sample Households People % poor
Constructing scorecards 18,846 39,459,467 32.0
Associating scores with likelihoods 9,665 20,407,790 31.4
Testing accuracy 9,503 19,760,021 31.9
Source: 2002 APIS. 38,014 79,627,278 31.8

34
Figure 2: Official poverty lines and “half” poverty lines, pesos/person/year
Official line "Half" line Official line "Half" line Official line "Half" line Official line "Half" line
Province Urban Rural Urban Rural Province Urban Rural Urban Rural Province Urban Rural Urban Rural Province Urban Rural Urban Rural
NCR Region IV Region VIII Region XII
1st District 16,496 N/A 13,353 N/A Batangas 15,993 15,002 11,410 11,230 Eastern Samar 10,617 9,690 6,658 6,461 Lanao del Norte 12,393 11,630 8,014 6,336
2nd District 16,007 N/A 12,875 N/A Cavite 14,851 16,240 11,992 12,480 Leyte 10,639 10,460 7,978 6,473 North Cotabato 11,172 9,761 7,680 6,185
3rd District 15,256 N/A 12,739 N/A Laguna 14,147 12,312 11,480 9,941 Northern Samar 9,726 9,503 6,751 6,045 Sultan Kudarat 11,940 10,565 7,444 6,893
4th District 16,654 N/A 13,828 N/A Marinduque 12,301 11,639 8,726 8,001 Western Samar 10,868 10,523 8,320 6,799
Occidental Mindoro 12,271 12,327 8,409 7,420 Southern Leyte 11,033 9,921 8,515 6,795
Oriental Mindoro 15,095 13,938 10,368 8,859 Biliran 10,218 10,644 6,816 7,375
Palawan 13,541 10,729 9,017 7,520
Quezon 13,430 12,605 9,478 8,318
Rizal 14,264 13,561 10,870 10,192
Romblon 12,770 11,234 7,267 7,006
Aurora 12,121 11,469 7,304 7,379

Region I Region V Region IX CAR

Ilocos Norte 13,175 13,688 9,000 9,271 Albay 15,239 11,763 10,251 6,954 Basilan 11,891 9,350 8,984 7,604 Abra 13,201 13,928 10,075 7,987
Ilocos Sur 12,768 14,368 9,746 8,375 Camarines Norte 13,931 11,259 8,822 6,686 Zamboanga del Norte 11,715 9,377 7,440 5,731 Benguet 15,300 13,309 12,976 8,363
La Union 13,415 13,183 10,118 7,667 Camarines Sur 13,049 10,389 8,349 6,206 Zamboanga del Sur 10,676 9,385 7,374 4,772 Ifugao 13,353 12,330 8,581 5,780
Pangasinan 13,449 12,737 9,699 8,034 Catanduanes 13,523 10,653 9,541 6,292 Kalinga 12,128 11,469 8,070 6,688
Masbate 13,784 10,903 8,367 7,462 Mt. Province 17,044 14,863 9,953 7,033
Sorsogon 13,551 11,264 8,098 7,366 Apayao 11,030 11,200 6,688 7,593

Region II Region VI Region X ARMM

Batanes 15,490 12,386 13,525 8,522 Aklan 12,581 11,938 9,420 7,080 Bukidnon 11,125 9,649 6,890 6,140 Lanao del Sur 13,459 14,725 8,831 10,017
Cagayan 12,507 10,127 9,029 7,409 Antique 11,981 10,969 7,852 7,165 Camiguin 14,228 11,943 8,854 6,959 Maguindanao 14,247 11,996 9,427 7,022
Isabela 14,883 11,317 10,364 7,500 Capiz 12,354 10,781 8,280 8,446 Misamis Occidental 11,898 10,081 7,543 6,535 Sulu 13,487 12,602 7,054 7,684
Nueva Vizcaya 13,707 10,730 10,112 8,412 Iloilo 12,948 12,328 9,750 8,110 Misamis Oriental 12,649 11,508 8,121 7,239 Tawi-tawi 13,192 13,259 11,057 10,097
Quirino 12,072 10,670 9,464 6,885 Negros Occidental 11,507 11,463 8,057 9,345
Guimaras 12,293 11,469 7,379 8,443

Region III Region VII Region XI Others

Bataan 13,344 11,706 10,411 10,200 Bohol 11,070 10,060 6,754 5,903 Davao del Norte 11,648 11,401 8,596 7,551 Agusan del Norte 12,767 10,594 7,839 5,908
Bulacan 14,822 13,265 13,526 8,521 Cebu 10,950 9,817 8,100 6,498 Davao del Sur 12,457 9,912 8,750 6,529 Agusan del Sur 12,355 11,104 8,433 6,000
Nueva Ecija 16,048 14,182 11,943 10,156 Negros Oriental 11,587 8,358 8,057 8,345 Davao Oriental 12,624 10,289 10,089 7,652 Surigao del Norte 13,813 11,261 8,456 6,050
Pampanga 15,459 14,111 12,068 11,631 Siquijor 11,823 9,361 7,283 5,553 South Cotabato 12,803 11,659 8,140 6,300 Surigao del Sur 12,422 10,694 8,356 6,663
Tarlac 13994 12409 10099 8675 Saranggani 12,674 11,719 7,482 4,944
Source: National Statistic Coordination Board (2004) and calculations based on the 2002 APIS.

35
Figure 3: Poverty indicators ranked by uncertainty coefficient
Uncertainty
coefficient Indicator (Value for those most likely “poor”; Value for those least likely “poor”)
1. 247 During the past six months, how much on average did the household spend per month on fuel, light, and water (charcoal,
firewood, LPG, kerosene/gas, electricity, candles, oils, water, etc.) (<P83/person; ≥P83/person)
2. 214 During the past six months, how much on average did the household spend per week on meat and meat preparations (fresh
chicken, beef, pork, carabeef, goat’s meat, corned beef, luncheon meat, meatloaf, Vienna sausage, longaniza, chorizo, hot
dogs, tocino, tapa, etc.) (<P19/person; ≥P19/person)
3. 214 During the past six months, how much on average did the household spend per month on personal care and effects (cleansing
cream, body deodorant, lotion, baby oil, toilet/bath soap, tissue paper, toothpaste, sanitary napkins, shampoo, jewelry,
handbags, wallets, wristwatches, haircuts, manicure or pedicure, etc.) (<P30/person; ≥P30/person)
4. 195 During the past six months, how much on average did the household spend per month on transportation and communication
(bus, jeepney, tricycle, air or water transport fare, gasoline/diesel, driver’s salary, telephone bills, postage stamps,
telegrams, driving lessons, feeds for animals used for transport, etc.) (<P38/person; ≥P38/person)
5. 190 Does the family own a gas stove or gas range? (No; Yes)
6. 183 Does the family own a refrigerator? (No; Yes)
7. 181 How many television sets does the family own? (0; 1; 2 or more)
8. 162 During the past six months, how much on average did the household spend per week on non-alcoholic beverages (soft drinks,
pineapple juice, orange juice, ice candy, ice drop, ice buko, etc.) (<P3/person; ≥P3/person)
9. 152 During the past six months, how much on average did the household spend per week on dairy products and eggs (milk, ice cream,
butter, cheese, fresh eggs, balut, salted eggs) (<P6/person; ≥P6/person)
10. 145 How many telephones and/or cell phones does the family own? (0; 1; 2 or more)
11. 140 During the past six months, how much on average did the household spend per week on food regularly consumed outside the
home (meals at school, place of work, restaurants, merienda or snacks, etc.) (<P11/person; ≥P11/person)
12. 136 Does the family own a sala set? (No; Yes)
13. 127 Does the family own a washing machine? (No; Yes)
14. 126 Is there any electricity in the building/house? (No; Yes)
15. 123 Does the family own a dining set? (No; Yes)
16. 121 During the past six months, how much did the household spend on clothing, footwear, and other wear (clothing and ready-made
apparel, footwear, sewing materials, accessories, service fees, etc.) (<P200/person; ≥P200/person)
17. 113 What is the highest grade completed by a household member? (Graduated secondary or less; 1 or more years of post-secondary)
18. 105 What is the primary occupation of the male head/spouse? (Farmers and laborers; Clerks, trades, special occupations, and
occupations not elsewhere classified; Clerks, and plant and machine operators; Technicians, and officials of government
and special-interest organizations; Professionals)

36
Figure 3 (cont.): Poverty indicators ranked by uncertainty coefficient
Uncertainty
coefficient Indicator (Value for those most likely “poor”; Value for those least likely “poor”)
19. 104 What is the house’s main source of water supply? (Spring, river, stream, dug well, or rain; Peddler or others; Own-use or shared-
use from a tubed/piped well or a community water system)
20. 102 Does the family have health insurance from Philhealth, a Health Maintenance Organization, a private health-insurance company,
or a community or cooperative? (No; Yes)
21. 98 Does the family have health insurance with Philhealth? (No; Yes)
22. 95 On a ladder with 10 steps going from lowest/poorest to highest/richest, on which step would you be? (1; 2; 3; 4; 5; 6 or more)
23. 94 How many people in the family are aged 0 to 17? (5 or more; 4; 2 or 3; 0 or 1)
24. 89 What are the house’s outer walls made of? (Salvaged or makeshift materials or all light or predominantly light materials such as
cogon, nipa, or sawali, bamboo, or anahaw; All strong or predominantly strong materials such as iron, aluminum, tile,
concrete, brick, stone, wood, or asbestos)
25. 86 What kind of toilet facility does the family have in the house? (None, open pit, or others; Closed pit; Water sealed)
26. 80 How many radios does the family own? (0; 1; 2 or more)
27. 78 What type of construction materials is the house’s roof made of? (Salvaged or makeshift materials or all light or predominantly
light materials such as cogon, nipa, or anahaw; All strong or predominantly strong materials such as galvanized iron,
aluminum tile, concrete, brick, stone, or asbestos)
28. 77 What is the highest grade completed by the male head/spouse? (Grades I to V of elementary; Did not complete any grade; Grade
VI of elementary to graduate of secondary; 1 or more years of post-secondary)
29. 77 How many people are there in the family from ages 0 to 17? (3 or more; 2; 1; 0)
30. 77 What is the highest grade completed by the female head/spouse? (I to V of elementary; None; VI of elementary to graduate of
secondary; 1 or more years of post-secondary)
31. 69 What is the primary occupation of the female head/spouse? (Farmers and laborers; Clerks, trades, special occupations, and
occupations not elsewhere classified; Clerks, and plant and machine operators; Technicians, and officials of government
and special-interest organizations; Professionals)
32. 61 Does the family live in an urban area? (No; Yes)
33. 58 How many children are there aged 17 or younger per adult aged 18 or older? (1 or more; <1)
34. 58 Does the family engage in crop farming or gardening? (Yes; No)
35. 53 Does the family own a vehicle? (No; Yes)
36. 48 What is the floor area of house in square meters? (50 or less; ≥50)
37. 47 Does the household have some type of health insurance? (No; Yes)
38. 41 During the past six months, did the household regularly consume food outside the home (meals at school, place of work,
restaurants, merienda or snacks, etc.)? (No; Yes)

37
Figure 3 (cont.): Poverty indicators ranked by uncertainty coefficient
Uncertainty
coefficient Indicator (Value for those most likely “poor”; Value for those least likely “poor”)
39. 40 During the past six months, did the household buy non-alcoholic beverages (soft drinks, pineapple juice, orange juice, ice
candy, ice drop, ice buko, etc.)? (No; Yes)
40. 39 How many children are there aged 11 or younger per adult aged 18 or older? (0.5 or more; <0.5)
41. 38 During the past six months, did the household make any deposits in banks? (No; Yes)
42. 35 During the past six months, did anyone in the family receive cash, gifts, support, or relief from abroad (including pensions
retirement, workmen’s compensation, dividends from investments, etc.)? (No; Yes)
43. 31 Is the male head/spouse self-employed? (No; Yes)
44. 22 During the past six months, did the household buy dairy products and eggs (milk, ice cream, butter, cheese, fresh eggs, balut,
salted eggs)? (No; Yes)
45. 22 Do all children in the family of ages 6 to 11 go to school? (No; Yes)
46. 22 Do any family members assist in the family business? (No; Yes)
47. 21 During the past six months, did anyone in the family make a deposit in a bank, receive interest on savings, or make a
withdrawal from a savings account? (No; Yes)
48. 21 Do any family members have salaried employment? (No; Yes)
49. 18 In the next 12 months, do you expect your household’s economic conditions to worsen, stay the same, or improve? (Worsen;
Stay the Same; Improve)
50. 17 Does the family have any income from entrepreneurial activities? (Yes; No)
51. 16 Is your household’s situation at present compared with the last 12 months worse, about the same, or better? (Worse, About the
same, Better)
52. 16 Are any household members involved with self-employment? (No; Yes)
Source: Based on 2002 APIS.

38
Figure 4: A simple poverty scorecard for the Philippines
Indicator Values Points
1. How many people in the family are aged 0 to 5 or more 3 or 4 1 or 2 Zero
17? 0 7 16 27
2. Does the family own a gas stove or gas range? No Yes
0 13
3. How many television sets does the family own? Zero 1 2 or more
0 9 18
4. What are the house's outer walls made of? Light (cogon , nipa , or sawali , bamboo, Strong (iron, aluminum,
anahaw ) tile, concrete, brick, stone,
wood, asbestos)
0 4
5. How many radios does the family own? Zero 1 2 or more
0 3 10
6. Does the family own a sala set? No Yes
0 9
7. What is the house's roof made of? Light (Salvaged, makeshift, cogon , nipa , or Strong (Galvanized iron,
anahaw ) aluminum tile, concrete,
brick, stone, or asbestos)
0 2
8. What kind of toilet facility does the family have? None, open pit, closed pit, or other Water sealed
0 3
9. Do all children in the family of ages 6 to 11 No Yes No children ages 6-11
go to school? 0 4 6
10. Do any family members have salaried employment? No Yes
0 6
Source: Calculations based on the 2002 APIS. Total:

39
Figure 5: Scores and poverty likelihoods
Poverty likelihood % of people % of people
for people with <=score >score
Score score in range (%) who are poor who are non-poor
0-4 99.3 99.3 68.4
5-9 92.5 94.6 69.4
10-14 91.9 93.2 70.9
15-19 93.4 93.3 72.5
20-24 77.6 88.3 74.3
25-29 76.8 84.8 77.0
30-34 77.7 82.9 80.6
35-39 48.6 74.0 83.4
40-44 48.3 68.7 86.8
45-49 33.6 63.0 89.1
50-54 34.4 58.4 92.7
55-59 22.6 55.5 94.0
60-64 10.1 49.0 94.9
65-69 10.2 43.7 96.6
70-74 6.9 40.1 97.9
75-79 3.8 37.1 98.8
80-84 2.1 34.1 100.0
85-89 0.0 32.9 100.0
90-94 0.0 32.5 100.0
95-100 0.0 32.1 100.0
Surveyed cases weighted to represent the Filipino population.
Source: Based on the 2002 APIS.

40
Figure 6: Confidence intervals for estimated poverty likelihoods associated
100
95 Upper 99% bound
Poverty likelihood (mean, 90-, 95-, 99-percent two-sided

90 Upper 95% bound

85 Upper 90% bound
80 Mean
75 Lower 90% bound
70 Lower 95% bound
65
confidence intervals)

Lower 99% bound

60
55
50
45
40
35
30
25
20 Average confidence intervals 90: 2.5
15 across scores, weighted by 95: 3.0
10 persons (+/- percentage points): 99: 3.9
5
0
0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95-
100
Range of score

41
Figure 7: Differences between estimated and true poverty likelihoods
20
18 Upper 99% bound
Upper 95% bound
Difference between predicted and estimated

16
Upper 90% bound
14
Mean
12
Lower 90% bound
10
Lower 95% bound
8
poverty likelihood

Lower 99% bound

6
4
2
0
-2
-4
-6
-8 Average confidence Mean: 3.6
-10 intervals across scores, 90: 2.5
-12 weighted by persons (+/- 95: 3.0
-14 percentage points): 99: 3.9
-16
0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95-
Range of score 100

42
Figure 8: ROC curve of ability to rank-order households by poverty status
1.00
Proportion of poor and non-poor with scores <=

0.95 "Poor"
0.90
0.85 Targeting the
a given score when ordered by score

0.80 lowest-scoring 20%

0.75 of households
0.70 targets 6% of the
0.65 non-poor and 51%
0.60 of the poor
0.55
0.50
0.45
0.40 Area KS
0.35 Random 0.0 0.0
0.30 Scorecard 75.5 59.2
0.25 Perfect 100 100
0.20
0.15
0.10
0.05 "Non-poor"
0.00
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Percentage of all cases, ordered by score

43
Figure 9: General classification matrix
Targeting segment
Targeted Non-targeted
A. B.
True poverty status

Truly poor Truly poor

Poor
correctly mistakenly
targeted non-targeted
C. D.
Truly non-poor Truly non-poor
Non-poor
mistakenly correctly
targeted non-targeted

44
Figure 10: People by targeting classification and score
A. B. C. D.
Truly poor Truly poor Truly non-poor Truly non-poor
correctly mistakenly mistakenly correctly
Score targeted non-targeted targeted non-targeted
0-4 0.7 31.4 0.0 67.9
5-9 2.1 30.0 0.1 67.8
10-14 4.3 27.8 0.3 67.6
15-19 6.4 25.6 0.5 67.5
20-24 9.0 23.1 1.2 66.7
25-29 12.4 19.7 2.2 65.7
30-34 16.6 15.5 3.4 64.5
35-39 20.0 12.1 7.0 60.9
40-44 23.4 8.7 10.6 57.3
45-49 25.6 6.5 15.0 52.9
50-54 28.3 3.8 20.2 47.8
55-59 29.2 2.8 23.4 44.5
60-64 30.1 2.0 31.3 36.7
65-69 31.1 1.0 40.1 27.8
70-74 31.6 0.4 47.2 20.7
75-79 31.9 0.2 54.0 14.0
80-84 32.1 0.0 62.0 5.9
85-89 32.1 0.0 65.5 2.5
90-94 32.1 0.0 66.6 1.3
95-100 32.1 0.0 67.9 0.0
Figures normalized to sum to 100.
Source: Based on the 2002 APIS.

45
Figure 11: General net-benefit matrix
Targeting segment
Targeted Non-targeted
True poverty status

Poor α β

Non-poor γ δ

46
Figure 12: “Total Accuracy” net-benefit matrix
Targeting segment
Targeted Non-targeted
True poverty status

Poor 1 0

Non-poor 0 1

47
Figure 13: Total net benefit for some common net-benefit matrices
Non-poverty
Total Accuracy Poverty Accuracy Accuracy Undercoverage Leakage
(A + B) 100*A / (A+B) 100*D / (C+D) 100*B / (A+B) 100*C / (A+C)
1 0 1 0 0 0 0 -1 0 0
Score 0 1 0 0 0 1 0 0 -1 0
0-4 68.6 2.2 100.0 97.8 0.8
5-9 69.9 6.6 99.8 93.4 5.4
10-14 71.9 13.4 99.5 86.6 6.8
15-19 73.9 20.1 99.3 79.9 6.7
20-24 75.7 27.9 98.3 72.1 11.7
25-29 78.1 38.6 96.7 61.4 15.2
30-34 81.1 51.7 95.0 48.3 17.1
35-39 80.9 62.3 89.7 37.7 26.0
40-44 80.7 72.8 84.3 27.2 31.3
45-49 78.5 79.7 77.9 20.3 37.0
50-54 76.0 88.2 70.3 11.8 41.6
55-59 73.7 91.1 65.5 8.9 44.5
60-64 66.8 93.9 54.0 6.1 51.0
65-69 58.9 97.0 41.0 3.0 56.3
70-74 52.3 98.6 30.5 1.4 59.9
75-79 45.9 99.5 20.5 0.5 62.9
80-84 38.0 100.0 8.7 0.0 65.9
85-89 34.5 100.0 3.6 0.0 67.1
90-94 33.4 100.0 1.9 0.0 67.5
95-100 32.1 100.0 0.0 0.0 67.9
All figures in percentage units.

48
Figure 14: “Poverty Accuracy” net-benefit matrix
Targeting segment
Targeted Non-targeted
True poverty status

Poor 1 0

Non-poor 0 0

49
Figure 15: Net-benefit matrix for BPAC
Targeting segment
Targeted Non-targeted
True poverty status

Poor 1 1

Non-poor -1 0

50
Figure 16: Poverty likelihoods for the very poor, poor, and non-poor
by score
Poverty likelihood in score range Share of cases <= score
Score Very Poor Poor Non-poor Very Poor Poor Non-poor
0-4 85.0 14.3 0.7 85.0 14.3 0.7
5-9 79.7 12.8 7.6 81.3 13.3 5.4
10-14 61.9 30.0 8.1 71.3 21.9 6.8
15-19 70.5 22.9 6.6 71.0 22.3 6.7
20-24 53.2 24.4 22.4 65.4 22.9 11.7
25-29 42.4 34.4 23.2 58.3 26.4 15.2
30-34 35.2 42.6 22.3 52.1 30.8 17.1
35-39 23.8 24.8 51.4 44.8 29.2 26.0
40-44 22.2 26.1 51.7 40.1 28.6 31.3
45-49 16.5 17.1 66.4 36.3 26.7 37.0
50-54 12.6 21.8 65.6 32.4 25.9 41.6
55-59 8.4 14.2 77.4 30.5 25.0 44.5
60-64 4.7 5.4 90.0 26.8 22.2 51.0
65-69 2.5 7.6 89.9 23.5 20.2 56.3
70-74 1.7 5.2 93.1 21.4 18.7 59.9
75-79 1.6 2.2 96.3 19.8 17.4 62.9
80-84 0.7 1.4 97.9 18.1 16.0 65.9
85-89 0.0 0.0 100.0 17.5 15.4 67.1
90-94 0.0 0.0 100.0 17.3 15.2 67.5
95-100 0.0 0.0 100.0 17.0 15.0 67.9
Total: 17.0 15.0 67.9

51
Figure 17: Confidence intervals for estimated poverty likelihoods for
being very poor associated with scores
100
Upper 99% bound
Difference between predicted and estimated

90
Upper 95% bound
Upper 90% bound
80
Mean
poverty likelihood (% pts.)

70 Lower 90% bound

Lower 95% bound
60 Lower 99% bound

50

40
Average confidence intervals 90: 2.2
30 across scores, weighted by 95: 2.8
persons (+/- percentage points): 99: 3.6
20

10

0
0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95-
Range of score 100

52
Figure 18: Differences between estimated and true poverty likelihoods
for the very poor
20
Upper 99% bound
Upper 95% bound
Difference between predicted and estimated

15 Upper 90% bound

Mean
10 Lower 90% bound
Lower 95% bound
poverty likelihood

Lower 99% bound

5

-5

Average confidence Mean: 1.8

-10 intervals across scores, 90: 2.5
weighted by persons (+/- 95: 3.0
-15 percentage points): 99: 4.0

-20
0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95-
Range of score 100

53
Figure 19: ROC curve of ability to rank-order households by very poor
versus not very poor poverty status
1.00
"Very poor"
Proportion of poor and non-poor with scores <=

0.95
0.90
0.85
a given score when ordered by score

0.80
0.75 Targeting the lowest-scoring
0.70 30% of households targets
0.65 21% of the not very poor and
0.60 77% of the very poor
0.55
0.50
0.45
0.40
0.35 Area KS
0.30 Random 0.0 0.0
0.25 Scorecard 73.4 56.0
0.20 Perfect 100 100
0.15
0.10
0.05 "Non-poor"
0.00
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Percentage of all cases, ordered by score

54
Figure 20: Classification matrix, three segments
Targeting segment
Very Poor Poor Non-poor
A. B. C.
Truly very poor Truly very poor Truly very poor
Very Poor correctly incorrectly incorrectly
True poverty status

targeted as very poor targeted as poor targeted as non-poor

D. E. F.
Truly poor Truly poor Truly poor
Poor incorrectly correctly incorrectly
targeted as very poor targeted as poor targeted as non-poor
G. H. I.
Truly non-poor Truly non-poor Truly non-poor
Non-poor incorrectly incorrectly correctly
targeted as very poor targeted as poor targeted as poor

55
Figure 21: Net-benefit matrix, three segments
Targeting segment
Very Poor Poor Non-poor

α β γ
Very Poor
True poverty status

δ ε ζ
Poor

η θ ι
Non-poor

56
Figure 22: Classification results, very poor/poor cut-offs from 0 to 44 and
poor/non-poor cut-offs from 5 to 49
Upper bound, poor segment
5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44
Upper bound, very poor segment

12 24 301 12 53 272 12 85 239 12 119 205 12 157 168 12 194 130 12 227 98 12 258 67
0-4 2 4 291 2 18 277 2 28 267 2 44 251 2 74 221 2 120 175 2 154 141 2 190 105
0 2 1,340 0 6 1,336 0 9 1,333 0 23 1,319 0 44 1,298 0 68 1,275 0 139 1,204 0 210 1,132
36 29 272 36 61 239 36 95 205 36 133 168 36 170 130 36 203 98 36 234 67
5-9 6 14 277 6 25 267 6 40 251 6 71 221 6 116 175 6 150 141 6 186 105
2 4 1,336 2 7 1,333 2 21 1,319 2 42 1,298 2 65 1,275 2 136 1,204 2 208 1,132
65 32 239 65 66 205 65 104 168 65 141 130 65 174 98 65 204 67
10-14 20 10 267 20 26 251 20 56 221 20 102 175 20 136 141 20 172 105
6 3 1,333 6 17 1,319 6 38 1,298 6 62 1,275 6 133 1,204 6 204 1,132
97 34 205 97 71 168 97 109 130 97 142 98 97 172 67
15-19 30 16 251 30 46 221 30 91 175 30 126 141 30 162 105
9 14 1,319 9 35 1,298 9 59 1,275 9 130 1,204 9 201 1,132
131 37 168 131 75 130 131 108 98 131 138 67
20-24 46 30 221 46 76 175 46 110 141 46 146 105
23 21 1,298 23 44 1,275 23 115 1,204 23 187 1,132
168 38 130 168 70 98 168 101 67
25-29 76 45 175 76 80 141 76 116 105
44 24 1,275 44 95 1,204 44 166 1,132
206 33 98 206 63 67
30-34 122 34 141 122 70 105
68 71 1,204 68 142 1,132
239 31 67
35-39 156 36 105
139 71 1,132

40-44

45-49

Figures in units of 10,000 people.

57
Figure 22 (cont.): Classification results, very poor/poor cut-offs from 0 to 49
and poor/non-poor cut-offs from 50 to 100
Upper bound, poor segment
50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89
12 299 26 12 306 19 12 314 11 12 319 6 12 321 3 12 324 1 12 325 0 12 325 0 12
0-4 2 246 49 2 258 37 2 267 28 2 282 13 2 290 5 2 293 2 2 295 0 2 295 0 2
0 399 944 0 463 880 0 618 724 0 793 550 0 933 409 0 1,066 276 0 1,226 117 0 1,293 49 0
36 275 26 36 282 19 36 290 11 36 295 6 36 297 3 36 300 1 36 301 0 36 301 0 36
5-9 6 243 49 6 254 37 6 263 28 6 278 13 6 286 5 6 289 2 6 291 0 6 291 0 6
2 396 944 2 460 880 2 616 724 2 790 550 2 931 409 2 1,064 276 2 1,223 117 2 1,291 49 2
65 246 26 65 253 19 65 261 11 65 266 6 65 268 3 65 270 1 65 272 0 65 272 0 65
10-14 20 228 49 20 240 37 20 249 28 20 264 13 20 272 5 20 275 2 20 277 0 20 277 0 20
6 393 944 6 456 880 6 612 724 6 786 550 6 927 409 6 1,060 276 6 1,220 117 6 1,287 49 6
97 214 26 97 220 19 97 229 11 97 233 6 97 236 3 97 238 1 97 239 0 97 239 0 97
15-19 30 218 49 30 230 37 30 239 28 30 254 13 30 262 5 30 265 2 30 267 0 30 267 0 30
9 390 944 9 453 880 9 609 724 9 783 550 9 924 409 9 1,057 276 9 1,217 117 9 1,284 49 9
131 180 26 131 186 19 131 195 11 131 199 6 131 202 3 131 204 1 131 205 0 131 205 0 131
20-24 46 202 49 46 214 37 46 223 28 46 238 13 46 246 5 46 249 2 46 251 0 46 251 0 46
23 375 944 23 439 880 23 594 724 23 769 550 23 910 409 23 1,043 276 23 1,202 117 23 1,270 49 23
168 142 26 168 149 19 168 157 11 168 162 6 168 165 3 168 167 1 168 168 0 168 168 0 168
25-29 76 172 49 76 184 37 76 193 28 76 208 13 76 216 5 76 219 2 76 221 0 76 221 0 76
44 355 944 44 419 880 44 574 724 44 749 550 44 890 409 44 1,022 276 44 1,182 117 44 1,250 49 44
206 105 26 206 111 19 206 120 11 206 124 6 206 127 3 206 129 1 206 130 0 206 130 0 206
30-34 122 127 49 122 138 37 122 147 28 122 162 13 122 170 5 122 173 2 122 175 0 122 175 0 122
68 331 944 68 395 880 68 550 724 68 725 550 68 866 409 68 999 276 68 1,158 117 68 1,226 49 68
239 72 26 239 79 19 239 87 11 239 92 6 239 94 3 239 97 1 239 98 0 239 98 0 239
35-39 156 92 49 156 104 37 156 113 28 156 128 13 156 136 5 156 139 2 156 141 0 156 141 0 156
139 260 944 139 324 880 139 479 724 139 654 550 139 795 409 139 928 276 139 1,087 117 139 1,155 49 139
269 41 26 269 48 19 269 56 11 269 61 6 269 64 3 269 66 1 269 67 0 269 67 0 269
40-44 192 56 49 192 68 37 192 77 28 192 92 13 192 100 5 192 103 2 192 105 0 192 105 0 192
210 189 944 210 253 880 210 408 724 210 583 550 210 723 409 210 856 276 210 1,016 117 210 1,083 49 210
291 20 26 291 27 19 291 35 11 291 40 6 291 42 3 291 45 1 291 46 0 291 46 0 291
45-49 214 34 49 214 46 37 214 55 28 214 70 13 214 78 5 214 81 2 214 83 0 214 83 0 214
296 103 944 296 166 880 296 322 724 296 496 550 296 637 409 296 770 276 296 930 117 296 997 49 296
Figures in units of 10,000 people.

58
Figure 22 (cont.): Classification results, very poor/poor cut-offs from 50 to 94
and poor/non-poor cut-offs from 55 to 100
55-59 60-64 65-69 70-74 75-79 80-84 85-89
311 7 19 311 15 11 311 20 6 311 23 3 311 25 1 311 26 0 311 26 0
Upper bound, very poor segment

50-54 248 12 37 248 21 28 248 36 13 248 44 5 248 47 2 248 49 0 248 49 0

399 64 880 399 219 724 399 394 550 399 535 409 399 668 276 399 827 117 399 895 49
317 8 11 317 13 6 317 16 3 317 18 1 317 19 0 317 19 0
55-59 260 9 28 260 24 13 260 32 5 260 35 2 260 37 0 260 37 0
463 155 724 463 330 550 463 471 409 463 604 276 463 763 117 463 831 49
326 5 6 326 8 3 326 10 1 326 11 0 326 11 0
60-64 269 15 13 269 23 5 269 26 2 269 28 0 269 28 0
618 175 550 618 316 409 618 449 276 618 608 117 618 676 49
330 3 3 330 5 1 330 6 0 330 6 0
65-69 284 8 5 284 11 2 284 13 0 284 13 0
793 141 409 793 274 276 793 433 117 793 501 49
333 2 1 333 3 0 333 3 0
70-74 292 3 2 292 5 0 292 5 0
934 133 276 934 292 117 934 360 49
335 1 0 335 1 0
75-79 295 2 0 295 2 0
1,066 159 117 1,066 227 49
336 0 0
80-84 297 0 0
1,226 68 49

85-89

90-94

Figures in units of 10,000 people.

59
Figure 23: Classification results, very poor 0–24,
poor 25–34, and non-poor 35–100
People with score in range
Segment Score Very Poor Poor Non-poor
0-4 12 2 0
5-9 24 4 2
Very poor 10-14 131 29 46 14 23 4
0-24 15-19 65% 32 23% 10 12% 3
20-24 34 16 14
Poor 25-29 75 37 76 30 44 21
25-34 30-34 38% 38 39% 45 23% 24
35-39 33 34 71
40-44 31 36 71
45-49 21 22 86
50-54 20 34 103
55-59 7 12 64
Non-poor 60-64 130 8 175 9 1,275 155
35-100 65-69 8% 5 11% 15 81% 175
70-74 3 8 141
75-79 2 3 133
80-84 1 2 159
85-89 0 0 68
90-94 0 0 23
95-100 0 0 26
Total: 336 297 1,342
Counts of people are in units of 10,000.

60
Figure 24: An example net-benefit matrix
reflecting common values
Targeting segment
Very Poor Poor Non-poor

+3 -2 -6
Very Poor
True poverty status

-1 +2 -2
Poor

-2 -1 +1
Non-poor

Note: This is an example. Each program should define its own net-benefit matrix.

61
Figure 25: Computation of total net benefit for a
cut-off pair of 20–24 and 30–34
Cell Persons Net benefit/person Net benefit
A. Truly very poor as very poor 131 +3 +393
B. Truly very poor as poor 75 –2 –150
C. Truly very poor as non-poor 130 –6 –780
D. Truly poor as very poor 46 –1 –46
E. Truly poor as poor 76 +2 +152
F. Truly poor as non-poor 175 –2 –350
G. Truly non-poor as very poor 23 –2 –46
H. Truly non-poor as poor 44 –1 –44
I. Truly non-poor as non-poor 1,275 +1 +1,275
Total net benefit: +404
Note: Persons are counted in units of 10,000.

62
Figure 26: Poverty likelihoods for the very poor, poor, $4/day poor,
and non-poor by score
Poverty likelihood in score range Share of cases <= score
Score Very Poor Poor Poor $4/day Non-poor Very Poor Poor Poor $4/day Non-poor
0-4 85.0 14.3 0.4 0.4 85.0 14.3 0.4 0.4
5-9 79.7 12.8 4.4 3.2 81.3 13.3 3.1 2.3
10-14 61.9 30.0 6.3 1.8 71.3 21.9 4.8 2.0
15-19 70.5 22.9 4.5 2.1 71.0 22.3 4.7 2.0
20-24 53.2 24.4 12.8 9.5 65.4 22.9 7.3 4.4
25-29 42.4 34.4 14.4 8.8 58.3 26.4 9.5 5.8
30-34 35.2 42.6 13.1 9.2 52.1 30.8 10.4 6.7
35-39 23.8 24.8 23.1 28.3 44.8 29.2 13.7 12.3
40-44 22.2 26.1 22.6 29.1 40.1 28.6 15.5 15.7
45-49 16.5 17.1 27.0 39.4 36.3 26.7 17.4 19.6
50-54 12.6 21.8 18.6 47.0 32.4 25.9 17.6 24.1
55-59 8.4 14.2 19.8 57.7 30.5 25.0 17.8 26.7
60-64 4.7 5.4 20.6 69.4 26.8 22.2 18.2 32.8
65-69 2.5 7.6 14.5 75.4 23.5 20.2 17.6 38.7
70-74 1.7 5.2 11.8 81.2 21.4 18.7 17.1 42.8
75-79 1.6 2.2 4.5 91.7 19.8 17.4 16.1 46.8
80-84 0.7 1.4 2.4 95.5 18.1 16.0 14.9 51.1
85-89 0.0 0.0 0.8 99.2 17.5 15.4 14.4 52.7
90-94 0.0 0.0 0.2 99.8 17.3 15.2 14.2 53.3
95-100 0.0 0.0 0.7 99.3 17.0 15.0 14.0 53.9
Total: 17.0 15.0 14.0 53.9

63
Figure 27: Confidence intervals for estimated poverty likelihoods for
being $4/day-or-less poor associated with scores
100
95 Upper 99% bound
Poverty likelihood (mean, 90-, 95-, 99-percent

90 Upper 95% bound

85 Upper 90% bound
80 Mean
two-sided confidence intervals)

75 Lower 90% bound

70 Lower 95% bound
65 Lower 99% bound
60
55
50
45
40
35
30
25
20 Average confidence intervals 90: 5.2
15 across scores, weighted by 95: 6.3
10 persons (+/- percentage points): 99: 8.3
5
0
0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95-
Range of score 100

64
Figure 28: Differences between estimated and true poverty likelihoods
for the $4/day-or-less poor
18
16
Difference between predicted and estimated

14
12
10
8
poverty likelihood

6
4
2
0
-2
-4
-6
-8
-10
-12
-14
-16
0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95-
Range of score 100

65
Figure 29: ROC curve of ability to rank-order households by
$4/day-or-less poverty status
1.00
Proportion of poor and non-poor with scores <=

0.95
0.90
0.85 Targeting the
a given score when ordered by score

0.80 lowest-scoring
0.75 20% of households
0.70 targets 3% of the
0.65 non-poor and 44%
0.60 of the $4/day poor
0.55
0.50
0.45
0.40 Area KS
0.35 Random 0.0 0.0
0.30 Scorecard 74.3 58.0
0.25 Perfect 100 100
0.20
0.15
0.10
0.05
0.00
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Percentage of all cases, ordered by score

66