PPM-V1 - Manual

PPM-V1 • Power and Performance Measure - Verbal Comprehension

J. Barrett
Hogrefe Ltd. The Test People, Oxford

Overview

The PPM series

The Power and Performance Measures (PPM) comprise nine aptitude and ability tests designed to meet the needs of occupational test users. The PPM-V1 belongs to this series. It is natural to describe all nine tests together, and this will be done here (with a few added particulars relating to the PPM-V1).

The PPM tests address two main elements of intellectual capacity: verbal ability and non-verbal ability. Verbal ability can be defined as understanding and reasoning abilities applied to commonly shared modes of written or spoken communication. This broad definition encompasses both language and numeracy comprehension and reasoning skills. Non–verbal ability incorporates problem solving skills that are not dependent on language or numeracy understanding. With this definition, the PPM tests that address these elements of ability are:

Verbal ability: Verbal Comprehension (V1), Verbal Reasoning (V2), Numerical Computation (N1), Numerical Reasoning (N2)

Non-verbal ability: Perceptual Reasoning (P1), Applied Power (A1), Spatial Ability (S1), Mechanical Understanding (M1), Processing Speed (C1)

A second distinction commonly made in the measurement of intellectual output is between ability and aptitude. Ability relates to what an individual can do now and may be able to do after experience and training; ability is specifically related to application of learned skills and previously attained knowledge to highly specified problems. Aptitude relates to a person's potential to acquire an ability; hence aptitude tests are less likely to measure elements of attained knowledge. This distinction is important when looking at the way in which the PPM works. An individual’s potential or aptitude is measured by the Power tests in the PPM series and ability is measured by the Performance tests. The inference is made and substantiated by studies of predictive validity that a person with a higher score on a test of aptitude will take less time to acquire ability than a person with a lower score.

Performance, as construed for the PPM tests, relates more strongly, though not inevitably, to the presence of prior knowledge. An example might be the Mechanical Understanding test (M1). Although it is important to note the value of this test for performance prediction, it can be the case that some people with no prior experience or knowledge occasionally achieve high scores. This is a function of elevated general ability overlapping into the domain of a specific ability test.

The Performance tests assess actual performance or ability, i.e. what an individual can do now, given the individual's experience or training. However, the component of prior knowledge should not be overly emphasised. It is found that the tests are practical and effective predictors of how well people will perform academically and vocationally. To put this another way, a person who scores as well as a qualified individual will have the potential to attain the equivalent applied knowledge.

Power, in respect of the PPM series, means reasoning, where prior knowledge contributes minimally; an example would be the Verbal Reasoning test (V2) where the vocabulary component is set at a low level so the test can be completed successfully by those with lower English vocabulary.

The Power tests within the full series of the PPM are the more abstract of the tests and measure ways of reasoning which do not change over time. Three of the Power tests measure numerical, verbal and perceptual reasoning, established as critical intellectual abilities for most roles. The fourth, the Applied Power test, is included because it often establishes certain individuals as having elevated reasoning ability who, while doing well on the abstract test, do poorly on the verbal, numerical and perceptual tests. Such individuals often do well in areas of work which do not require specific academic attainments.

The classification of PPM tests based on the principles above is as follows:

Table 1: Classification of PPM tests

Verbal Non-verbal

Power Verbal Reasoning (V2)
Numerical Reasoning (N2)
Perceptual Reasoning (P1)
Applied Power (A1)

Performance Verbal Comprehension (V1)
Numerical Computation (N1) Spatial Ability (S1)
Mechanical Understanding (M1)
Processing Speed (C1)

The range of PPM tests gives the test user valuable flexibility. This is especially the case in selection where the intellectual ability needs for the role are known. In these instances it is expected that particular tests from the PPM series will be combined for tailored testing that targets critical intellectual strengths. For example, certain tests can be combined for a technical context, others can be combined for a clerical context, and certain tests from the reasoning, verbal and numerical tests can be combined for a high level context. Finally, the complete nine test series is ideal for vocational guidance where meaningful differences between abilities need to be ascertained.

The individual PPM tests

The Power tests

Verbal Reasoning (V2): This test measures a person’s understanding of verbal concepts including comprehension and reasoning. The test deliberately keeps its level of vocabulary relatively low, as its purpose is to measure the way in which an individual reasons using verbal cues. The test is a useful indicator of a person's ability to understand and communicate. The test is again a logical one even though judgments have to be made in the absence of specific information. A correct choice has to be deduced or extrapolated, so providing some indication of critical thinking ability.

Numerical Reasoning (N2): The test measures a subject’s ability to reason with numbers. Each problem looks at a person’s ability to see the relationships between sets of numbers. The N2 is a good predictor of academic performance and also a strong indicator of success in jobs requiring an understanding of numerical data.

Perceptual Reasoning (P1): This is a reasoning test based upon non-language principles and in this respect differs from both the verbal and the numerical reasoning tests. As such it is more predictive for areas of science and technology and in common with the Applied Power test is a particularly good indicator of general intelligence and problem solving skills.

Applied Power (A1): This test measures the ability to apply reasoning to solving a series of logical problems. The test taker is presented with algebraic symbols which give cues as to what should be next in sequence. These cues are progressively separated so that the "size" of the logical problem increases in magnitude. The test requires an ability to reason from basic principles and depends not at all upon prior knowledge. The fact that the test combines letters and number for symbols is merely a convenience for publication. Diagrams and colours could just as easily be used.

The Performance tests

Verbal Comprehension (V1): This is a good academic predictor and a measure of attainment. It is a definite measure of comprehension because of the need to understand increasingly complex vocabulary. It should be pointed out, however, that people whose first language is not English may well achieve depressed scores.

Numerical Computation (N1): This test requires the application of the basic rules of arithmetic. Experience shows that people who really are numerate perform well on this task even though they have "forgotten" about mental arithmetic and particularly the division rule. All the problems are presented in the accepted mathematical sequence.

Spatial Ability (S1): The fundamental principle involved in this test is that the test taker must rotate and turn over shapes in the "mind's eye". The problems are of course presented two-dimensionally so that the candidate has to be able to "see" from the other side of the object or pattern. This is an essential ability whenever there is a need to visualize in three dimensions. As such it is a practical task, but the intellectual element required in rotating a number of hypotheses in order to form a single concept is also demanded.

Mechanical Understanding (M1): Basic principles of dynamics are presented in the form of mechanical devices or situations involving forces. As with the Spatial test, conceptual ability is sometimes different from physical dexterity - the two do not always go together. During the development of the PPM series, it was found that performance on this test is not as much affected by previous experience as is generally imagined and the test is therefore a reliable predictor of the test taker’s innate mechanical aptitude.

Processing Speed (C1): This is a speed test requiring accuracy in recognizing a stimulus and responding to it. At one level it can be regarded as a "clerical" test but otherwise it reveals an ability to order data systematically. Although the data in the test takes the form of English words, the test essentially looks at the way in which subjects process information. The lack of English as a first language is therefore likely to have less impact on test scores than is the case with the Verbal Comprehension test.

The scales in the PPM-V1

As well as returning the test score itself, Hogrefe TestSystem software can return the underlying numbers of correct, incorrect and missing answers.

Population-based norms; associated with items

Population-based norms; calculated

Derived/calculated score

(x) Dealing with missing values:
(0) ignore them
(1) use regression to estimate an answer
(2) use the middle scale point as answer
(3) use a defined score as answer

Characteristics Standard

Verbal Comprehension

Special advice

It is always recommended that decisions to include particular tests for any selection process should be based on the results of a job analysis or at least a detailed job description. If possible, a validation study might be conducted prior to using the tests for selection purposes. A validation study further ensures the appropriateness of a particular test's use and consequently increased fairness for applicants.

Area of application

For assessment in selection, development and career counselling programmes with individuals aged 18 to 60 years. The series can be administered individually or in a group setting.

*Table 1: Classification of PPM tests*
	Verbal	Non-verbal
Power	Verbal Reasoning (V2) Numerical Reasoning (N2)	Perceptual Reasoning (P1) Applied Power (A1)
Performance	Verbal Comprehension (V1) Numerical Computation (N1)	Spatial Ability (S1) Mechanical Understanding (M1) Processing Speed (C1)

Characteristics	Standard
Verbal Comprehension

Norms

The normative samples for the PPM-V1 are summarised in Table 3a:

Norm Standard

General Population

Age groups Neither gender- nor medium-specific

up to 80 yrs. N =

Working Population

Age groups Neither gender- nor medium-specific

up to 80 yrs. N =

Professional Manager Population

Age groups Neither gender- nor medium-specific

up to 80 yrs. N =

The largest normative sample, labelled 'British general population', consists of 1600 cases aged from 18 - 60. The sample came from across Britain. It comprised male and female schoolchildren, students, working adults and professionals from various fields.

Table 3b: PPM series normative sample sizes

Norm Group N

British General Population 1600

British Working Population various

Professional Managers 337

*Table 3b: PPM series normative sample sizes*
Norm Group	N
British General Population	1600
British Working Population	various
Professional Managers	337

Psychometric properties

Reliability

Internal Consistency

Reliability in psychometric tests can be defined as the proportion of the total variance in scores which is due to the true score rather than to other extraneous factors. In practice, a test's reliability is assessed through its consistency.

Internal consistency concerns the equivalence of parts of the test for measurement purposes. High internal consistency will ensure that the items within the test measure the same psychological construct. The best measure of the internal consistency of a test is Cronbach's Alpha. As a generalised rule of thumb we would expect an Alpha score of between 0.7 and 1.0. Cronbach's Alpha scores are provided in Table 4 below:

Table 4: Cronbach's Alpha (various samples, see appendix)
PPM test Cronbach's Alpha

(Power) Applied Power (A1) .88

Numerical Reasoning (N2) .83

Perceptual Reasoning (P1) .74

Verbal Reasoning (V2) .83

(Performance) Verbal Comprehension (V1) .84

Numerical Computation (N1) .86

Spatial Ability (S1) .79

Mechanical Understanding (M1) .79

Processing Speed (C1) .89

Inter-correlations between PPM tests

High inter-correlations between tests would indicate redundancy due to overlap in measurement. It is important that each test is measuring something different as otherwise separate tests will be measuring the same sphere of intelligence. The table below shows the correlations between each of the tests.

Table 5: Test correlations for the sample of 337 professional managers.

A1 C1 M1 N1 N2 P1 S1 V1 V2

Mean 13.07 28.84 18.39 21.16 12.88 12.57 11.46 40.57 17.12

SD 4.69 7.43 5.76 6.69 4.37 3.23 3.95 9.09 4.44

Applied Power (A1) *

Processing Speed (C1) 0.44 *

Mechanical Understanding (M1) 0.36 0.20 *

Numerical Computation (N1) 0.36 0.46 0.14 *

Numerical Reasoning (N2) 0.47 0.48 0.27 0.59 *

Perceptual Reasoning (P1) 0.46 0.44 0.32 0.40 0.51 *

Spatial Ability (S1) 0.49 0.34 0.45 0.30 0.51 0.43 *

Verbal Comprehension (V1) 0.25 0.28 0.33 0.24 0.25 0.15 0.25 *

Verbal Reasoning (V2) 0.46 0.53 0.22 0.37 0.43 0.45 0.36 0.34 *

It is clear from a visual inspection of these data that each test is measuring something different but it is important to be mindful of Bartram (1992) which advises that it is worthwhile ascertaining just how objectively reliable the differences between the tests are.

Reliability of the difference between tests

The formula for obtaining the reliability of the differences between tests where the scale score of each test has a different variation is given in Cronbach (ibid p279)

r(atrue-btrue) = (Va r_aa + Vb r_bb - 2Cab) / (Va + Vb - 2Cab)

where V is the variance, C the covariance, and r_aa and r_bb the reliabilities.

As the variance is sd^2 [sd squared, "^2" denotes superscript 2] and the covariance is the product of the two sd values and the correlation r_ab, Bartram (1992) helpfully suggests the following version of the same formula where S is the standard deviation:

r(atrue-btrue) = ( (Sa)^2 raa + (Sb)^2 rbb – 2 r_ab Sa Sb) / ( (Sa)^2 + (Sb)^2 – 2 r_ab Sa Sb)

This formula applied to the correlations in Table 5 gives the values shown in Table 6.

Table 6: Inter-scale reliabilities
A1 C1 M1 N1 N2 P1 S1 V1 V2

Applied Power (A1)

Processing Speed (C1) .901

Mechanical Understanding (M1) .907 .926

Numerical Computation (N1) .909 .889 .93

Numerical Reasoning (N2) .887 .896 .919 .869

Perceptual Reasoning (P1) .894 .912 .917 .913 .883

Spatial Ability (S1) .884 .916 .897 .919 .878 .896

Verbal Comprehension (V1) .925 .917 .914 .922 .925 .934 .927

Verbal Reasoning (V2) .889 .887 .924 .909 .895 .895 .906 .918

Validity

Evidence of validity is derived from review of the structure of the test by means of multivariate analyses. Correlations with other assessment instruments (NIIP, EAS and Bennett Mechanical tests) indicate convergent validity.

Table 7. PPM correlations with the NIIP Engineering Selection Test Battery

PPM test Number of participants NIIP test Correlation

Mechanical Understanding (M1) 157 Bennett Mechanical 0.49

Numerical Computation (N1) 47 Arithmetic (EA4) 0.55

Numerical Reasoning (N2) 47 Arithmetic (EA4) 0.35

Spatial Ability (S1) 47 Spatial test (82) 0.75

Spatial Ability (S1) 47 Bennett Mechanical 0.54

Verbal Comprehension (V1) 47 Verbal NIIP (90B) 0.61

Table 8. PPM correlations with the Employee Aptitude Survey (EAS)

PPM test Number of participants EAS test Correlation

Applied Power (A1) 144 Numerical Reasoning 0.44

Processing Speed (C1) 152 Visual Speed 0.74

Mechanical Understanding (M1) 158 Space Visualisation 0.53

Numerical Computation (N1) 20 Numerical 0.68

Numerical Reasoning (N2) 20 Numerical Reasoning 0.60

Spatial Ability (S1) 153 Space Visualisation 0.61

Perceptual Reasoning (P1) 158 Visual Pursuit 0.34

Perceptual Reasoning (P1) 140 Space Visualisation 0.53

Perceptual Reasoning (P1) 140 Symbolic Reasoning 0.49

Verbal comprehension (V1) 131 Word Knowledge 0.76

Verbal Reasoning (V2) 201 Verbal Reasoning 0.57

Table 9. PPM correlations with SHL tests - NIT2, NMG2, VMG2

PPM test Number of participants SHL test Correlation

Applied Power (A1) 107 NIT2 0.2938 (P=0.002)

Applied Power (A1) 41 NMG2 0.3893 (P=0.012)

Applied Power (A1) 148 VMG2 0.2310 (P=0.005)

Numerical Reasoning (N2) 182 NMG2 0.6017 (P=0.000)

Numerical Reasoning (N2) 182 VMG2 0.2928 (P=0.000)

Verbal Reasoning (V2) 97 NIT2 0.4551 (P=0.000)

Verbal Reasoning (V2) 100 NMG2 0.2575 (P=0.010)

Verbal Reasoning (V2) 197 VMG2 0.4386 (P=0.000)

NMG2 is a numerical critical reasoning test, VMG2 is a verbal critical reasoning test and NIT2 is also a numerical reasoning test but differs in item construction from the NMG2. All are published by SHL.

The correlation tables above provide us with the information to interpret the similarity of the abilities or aptitudes being measured by the tests. Higher correlations identify that the two tests being compared are more similar. Although this is a relative measure, Cohen (1988) notes that correlations with values between 0.3 and 0.5 indicate a medium similarity. This is what we would expect for tests that measure the same ability or aptitude but in different ways. We can see that the correlations given in the tables support the PPM series measuring the constructs for which they were designed.

*Table 4: Cronbach's Alpha (various samples, see appendix)*
	PPM test	Cronbach's Alpha
(Power)	Applied Power (A1)	.88
Numerical Reasoning (N2)	.83
Perceptual Reasoning (P1)	.74
Verbal Reasoning (V2)	.83
(Performance)	Verbal Comprehension (V1)	.84
Numerical Computation (N1)	.86
Spatial Ability (S1)	.79
Mechanical Understanding (M1)	.79
Processing Speed (C1)	.89

*Table 5: Test correlations for the sample of 337 professional managers.*
	A1	C1	M1	N1	N2	P1	S1	V1	V2
Mean	13.07	28.84	18.39	21.16	12.88	12.57	11.46	40.57	17.12
SD	4.69	7.43	5.76	6.69	4.37	3.23	3.95	9.09	4.44
Applied Power (A1)	*
Processing Speed (C1)	0.44	*
Mechanical Understanding (M1)	0.36	0.20	*
Numerical Computation (N1)	0.36	0.46	0.14	*
Numerical Reasoning (N2)	0.47	0.48	0.27	0.59	*
Perceptual Reasoning (P1)	0.46	0.44	0.32	0.40	0.51	*
Spatial Ability (S1)	0.49	0.34	0.45	0.30	0.51	0.43	*
Verbal Comprehension (V1)	0.25	0.28	0.33	0.24	0.25	0.15	0.25	*
Verbal Reasoning (V2)	0.46	0.53	0.22	0.37	0.43	0.45	0.36	0.34	*

*Table 6: Inter-scale reliabilities*
	A1	C1	M1	N1	N2	P1	S1	V1
Applied Power (A1)
Processing Speed (C1)	.901
Mechanical Understanding (M1)	.907	.926
Numerical Computation (N1)	.909	.889	.93
Numerical Reasoning (N2)	.887	.896	.919	.869
Perceptual Reasoning (P1)	.894	.912	.917	.913	.883
Spatial Ability (S1)	.884	.916	.897	.919	.878	.896
Verbal Comprehension (V1)	.925	.917	.914	.922	.925	.934	.927
Verbal Reasoning (V2)	.889	.887	.924	.909	.895	.895	.906	.918

*Table 7. PPM correlations with the NIIP Engineering Selection Test Battery*
PPM test	Number of participants	NIIP test	Correlation
Mechanical Understanding (M1)	157	Bennett Mechanical	0.49
Numerical Computation (N1)	47	Arithmetic (EA4)	0.55
Numerical Reasoning (N2)	47	Arithmetic (EA4)	0.35
Spatial Ability (S1)	47	Spatial test (82)	0.75
Spatial Ability (S1)	47	Bennett Mechanical	0.54
Verbal Comprehension (V1)	47	Verbal NIIP (90B)	0.61

*Table 8. PPM correlations with the Employee Aptitude Survey (EAS)*
PPM test	Number of participants	EAS test	Correlation
Applied Power (A1)	144	Numerical Reasoning	0.44
Processing Speed (C1)	152	Visual Speed	0.74
Mechanical Understanding (M1)	158	Space Visualisation	0.53
Numerical Computation (N1)	20	Numerical	0.68
Numerical Reasoning (N2)	20	Numerical Reasoning	0.60
Spatial Ability (S1)	153	Space Visualisation	0.61
Perceptual Reasoning (P1)	158	Visual Pursuit	0.34
Perceptual Reasoning (P1)	140	Space Visualisation	0.53
Perceptual Reasoning (P1)	140	Symbolic Reasoning	0.49
Verbal comprehension (V1)	131	Word Knowledge	0.76
Verbal Reasoning (V2)	201	Verbal Reasoning	0.57

*Table 9. PPM correlations with SHL tests - NIT2, NMG2, VMG2*
PPM test	Number of participants	SHL test	Correlation
Applied Power (A1)	107	NIT2	0.2938 (P=0.002)
Applied Power (A1)	41	NMG2	0.3893 (P=0.012)
Applied Power (A1)	148	VMG2	0.2310 (P=0.005)
Numerical Reasoning (N2)	182	NMG2	0.6017 (P=0.000)
Numerical Reasoning (N2)	182	VMG2	0.2928 (P=0.000)
Verbal Reasoning (V2)	97	NIT2	0.4551 (P=0.000)
Verbal Reasoning (V2)	100	NMG2	0.2575 (P=0.010)
Verbal Reasoning (V2)	197	VMG2	0.4386 (P=0.000)

Duration

Duration information for all PPM tests is included in the next section, on Scoring. The duration specifics for the PPM-V1 are as follows:

Test form Duration, ca. No. of items

Standard 8.00 min 64

Durations are absolute, from start to end of the test process (incl. instruction phase etc).

Test form	Duration, ca.	No. of items
Standard		8.00 min	64
Durations are absolute, from start to end of the test process (incl. instruction phase etc).

Scoring

PPM test Number of items Time limit (minutes) Scoring

Applied Power (A1) 25 12 correct answers

Processing Speed (C1) 50 3 correct answers - 1/3 wrong answers

Mechanical Understanding (M1) 31 8 correct answers - 1/4 wrong answers

Numerical Computation (N1) 40 6 correct answers - 1/3 wrong answers

Numerical Reasoning (N2) 25 10 correct answers - 1/4 wrong answers

Perceptual Reasoning (P1) 26 6 correct answers - 1/3 wrong answers

Spatial Ability (S1) 26 6 correct answers - 1/6 wrong answers

Verbal Comprehension (V1) 60 6 correct answers - 1/3 wrong answers

Verbal Reasoning (V2) 31 10 correct answers - 1/4 wrong answers

Note: In the scoring calculation, any fractional part arising from the division is discarded before the subtraction.

PPM test	Number of items	Time limit (minutes)	Scoring
Applied Power (A1)	25	12	correct answers
Processing Speed (C1)	50	3	correct answers - 1/3 wrong answers
Mechanical Understanding (M1)	31	8	correct answers - 1/4 wrong answers
Numerical Computation (N1)	40	6	correct answers - 1/3 wrong answers
Numerical Reasoning (N2)	25	10	correct answers - 1/4 wrong answers
Perceptual Reasoning (P1)	26	6	correct answers - 1/3 wrong answers
Spatial Ability (S1)	26	6	correct answers - 1/6 wrong answers
Verbal Comprehension (V1)	60	6	correct answers - 1/3 wrong answers
Verbal Reasoning (V2)	31	10	correct answers - 1/4 wrong answers

Special features of the computer version

In addition to the implementation on Hogrefe TestSystem (HTS), the PPM-V1 continues to be available as a paper-pencil test and on other platforms supplied through Hogrefe Ltd. The implementation on Hogrefe TestSystem provides the test user with additional insight into the result: in addition to obtaining the standard report with the normed test score, the test user equipped with HTS software can inspect the underlying numbers of correct, incorrect and missing responses, the individual item responses and even the item response latencies.

Appendix

Below we give some details of the reliability study with various samples of 'British working population'.

Applied Power (A1)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 384

EDUCATION : GCSE, ‘O’ Level, ‘A’ Level (mostly ‘O’ level)

GROUP DESCRIPTION : Clerical workers with large financial institutions

AGE RANGE : 17 Years to 37 Years (Median age = 22 Years) (Mean age = 22.26, N = 374)

SEX BREAKDOWN : Female 71 (18.5%) Male 313 (81.5%)

TEST STATISTICS

TEST SCORE : Mean = 11.82 ; SD = 4.39 ; N = 384 ; Score Range : 0 to 25

TEST RELIABILITY : Cronbach’s Alpha = 0.8771

CORRELATIONS : AGE WITH TEST SCORE : -0.0875 (N = 374)
* Not Significant

AGE DIFFERENCES : Up to and including 22 years : Mean = 12.2617 ; SD = 4.505 ; N = 214
23 Years upwards : Mean = 11.3313 ; SD = 4.297 ; N = 160
Mean Difference : 0.9304 ; 2-tail sign. = 0.040 ; Significant

SEX DIFFERENCES : Female : Mean = 12.7887 ; SD = 4.724 ; N = 71
Male : Mean = 11.6038 ; SD = 4.292 ; N = 313
Mean Difference : 1.1849 ; 2-tail sign. = 0.040 ; Significant

PROCESSING SPEED (C1)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 555

EDUCATION : GCSE, ‘O’ Level, ‘A’ Level (mostly ‘O’ Level)

GROUP DESCRIPTION : Clerical workers with large financial institutions

AGE RANGE : -

SEX BREAKDOWN : Female 425 (81.1%) Male 99 (18.9%)

TEST STATISTICS

TEST SCORE : Mean = 19.91 ; SD = 6.63 ; N = 555 ; Score Range : 0 to 42

TEST RELIABILITY : Cronbach’s Alpha = 0.8906

CORRELATIONS : AGE WITH TEST SCORE : -0.1647 (N = 225)
* Younger candidates score significantly higher

AGE DIFFERENCES : -

SEX DIFFERENCES : Female : Mean = 20.5388 ; SD = 6.414 ; N = 425
Male : Mean = 17.3838 ; SD = 7.132 ; N = 99
Mean Difference : -3.1550 ; 2-tail sign. = 0.000; Significant

MECHANICAL REASONING (M1)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 718

EDUCATION : No Formal Quals to Graduate (mostly ‘A’ level)

GROUP DESCRIPTION : Engineers, Engineering Tradespeople, Factory workers, manufacturing applicants, pilot applicants

AGE RANGE : 17 Years to 60 Years (Median age = 23 Years) (Mean age = 26.83 ; N = 616)

SEX BREAKDOWN : Female 49 (5.6%) Male 659 (94.4%)

TEST STATISTICS

TEST SCORE : Mean = 15.09 ; SD = 5.78, N = 718 ; Score Range : 0 to 31

TEST RELIABILITY : Cronbach’s Alpha = 0.7868

CORRELATIONS : AGE WITH TEST SCORE : -0.2519 (N = 616)
* Younger candidates score significantly higher

AGE DIFFERENCES : Up to and including 23 years : Mean = 16.4343 ; SD = 5.450 ; N = 327
24 Years upwards : Mean = 14.4948 ; SD = 5.810 ; N = 289
Mean Difference : 1.9396 ; 2-tail sign. = 0.000 ; Significant

SEX DIFFERENCES : Female : Mean = 12.1429 ; SD = 5.817 ; N = 49
Male : Mean = 15.3171 ; SD = 5.743 ; N = 659
Mean Difference : -3.1743 ; 2-tail sign. = 0.000 ; Significant

NUMERICAL COMPUTATION (N1)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 793

EDUCATION : No Formal Quals to Graduate (mostly ‘A’ level)

GROUP DESCRIPTION : Engineers, Engineering Tradespeople, Pilot applicants, Factory workers, Manufacturing applicants

AGE RANGE : 17 Years to 60 Years (Median age = 24 Years) (Mean age = 27.32 ; N = 669)

SEX BREAKDOWN : Female 57 (4.4%) Male 724 (95.6%)

TEST STATISTICS

TEST SCORE : Mean = 13.80 ; SD = 6.27, N = 793 ; Score Range : 0 to 37

TEST RELIABILITY : Cronbach’s Alpha = 0.8613

CORRELATIONS : AGE WITH TEST SCORE : -0.1655 (N = 142)
* Younger candidates score significantly higher

AGE DIFFERENCES : Up to and including 23 years : Mean = 14.6425 ; SD = 5.809 ; N = 328
24 Years upwards : Mean = 13.7462 ; SD = 6.479 ; N = 341
Mean Difference : 0.8963 ; 2-tail sign. = 0.060 ; Not significant

SEX DIFFERENCES : Female : Mean = 13.0063 ; SD = 5.515 ; N = 57
Male : Mean = 13.8718 ; SD = 6.356 ; N = 724
Mean Difference : -0.8655 ; 2-tail sign. = 0.318; Not Significant

NUMERICAL REASONING (N2)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 368

EDUCATION : From ‘O’ Level to Graduate (mostly ‘A’ level)

GROUP DESCRIPTION : Engineers, Engineering tradesmen (fitters, electricians etc.), Managers, Pilot applicants, Factory workers, Manufacturing applicants

AGE RANGE : 17 Years to 54 Years (Median age = 23 Years) (Mean age = 25.38 ; N = 255)

SEX BREAKDOWN : Female 90 (25.6%) Male 262 (74.4%)

TEST STATISTICS

TEST SCORE : Mean = 10.89 ; SD = 4.98, N = 368 ; Score Range : 0 to 25

TEST RELIABILITY : Cronbach’s Alpha = 0.8251

CORRELATIONS : AGE WITH TEST SCORE : -0.3710 (N = 255)
*Younger candidates scored significantly higher.

AGE DIFFERENCES : Up to and including 22 years : Mean = 13.7869 ; SD = 4.484 ; N = 122
23 Years upwards : Mean = 10.7519 ; SD = 4.831 ; N = 133
Mean Difference : 3.0350 ; 2-tail sign. = 0.000 ; Significant

SEX DIFFERENCES : Female : Mean = 11.8333 ; SD = 4.327 ; N = 90
Male : Mean = 10.7557 ; SD = 5.159 ; N = 262
Mean Difference : 1.0776 ; 2-tail sign. = 0.076 ; Not significant

PERCEPTUAL REASONING (P1)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 283

EDUCATION : No formal quals to ‘A’ level (mostly ‘O’ level)

GROUP DESCRIPTION : Factory workers, manufacturing applicants, engineering trades people

AGE RANGE : 17 Years to 54 Years (Median age = 22 Years) (Mean age = 26.71 ; N = 159)

SEX BREAKDOWN : Female 20 (7.5%) Male 247 (92.5%)

TEST STATISTICS

TEST SCORE : Mean = 11.32 ; SD = 3.65, N = 283 ; Score Range : 0 to 20

TEST RELIABILITY : Cronbach’s Alpha = 0.7387

CORRELATIONS : AGE WITH TEST SCORE : -0.2522 (N = 159)
* Younger candidates score significantly higher

AGE DIFFERENCES : Up to and including 21 years : Mean = 12.7805 ; SD = 3.410 ; N = 82
22 Years upwards : Mean = 11.2857 ; SD = 3.713 ; N = 77
Mean Difference : 1.4948 ; 2-tail sign. = 0.009 ; Significant

SEX DIFFERENCES : Female : Mean = 12.8000 ; SD = 3.636 ; N = 20
Male : Mean = 11.2753 ; SD = 3.587 ; N = 247
Mean Difference : 1.5247 ; 2-tail sign. = 0.069 ; Not significant

SPATIAL ABILITY (S1)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 590

EDUCATION : From ‘O’ level to Graduate (mostly ‘A’ level)

GROUP DESCRIPTION : Engineers, Engineering Trades people (fitters, electricians), pilot applicants, creatives

AGE RANGE : 16 Years to 59 Years (Median age = 23 Years) (Mean age = 25.66 ; N = 495)

SEX BREAKDOWN : Female 32 (5.4%) Male 558 (94.6%)

TEST STATISTICS

TEST SCORE : Mean = 9.27 ; SD = 4.15, N = 590 ; Score Range : 0 to 21

TEST RELIABILITY : Cronbach’s Alpha = 0.7891

CORRELATIONS : AGE WITH TEST SCORE : -0.2878 (N = 495)
* Younger candidates score significantly higher
* Younger candidates have higher educational level

AGE DIFFERENCES : Up to and including 23 years : Mean = 10.4148 ; SD = 3.719 ; N = 270
24 Years upwards : Mean = 8.1467 ; SD = 4.399 ; N = 225
Mean Difference : 2.2681 ; 2-tail sign. = 0.000 ; Significant

SEX DIFFERENCES : Female : Mean = 10.1875 ; SD = 4.083 ; N = 32
Male : Mean = 9.2133 ; SD = 4.147 ; N = 558
Mean Difference : 0.9742 ; 2-tail sign. = 0.196 ; Not significant

VERBAL COMPREHENSION (V1)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 162

EDUCATION : From ‘O’ level to Graduate (mostly ‘A’ level)

GROUP DESCRIPTION : Managers, engineering tradespeople

AGE RANGE : 18 Years to 60 Years (Median age = 21 Years) (Mean age = 26.25 ; N = 134)

SEX BREAKDOWN : Female 13 (8.7%) Male 137 (91.3%)

TEST STATISTICS

TEST SCORE : Mean = 21.20 ; SD = 10.79, N = 162 ; Score Range : 0 to 57

TEST RELIABILITY : Cronbach’s Alpha = 0.8380

CORRELATIONS : AGE WITH TEST SCORE : 0.2273 (N = 134)
* Older candidates score significantly higher

AGE DIFFERENCES : Up to and including 21 years : Mean = 17.3919 ; SD = 6.257 ; N = 74
22 Years upwards : Mean = 24.0833 ; SD = 12.395 ; N = 60
Mean Difference : -6.6914 ; 2-tail sign. = 0.000 ; Significant

SEX DIFFERENCES : Female : Mean = 22.1538 ; SD = 7.625 ; N = 13
Male : Mean = 20.3650 ; SD = 0.713 ; N = 137
Mean Difference : -1.7889 ; 2-tail sign. = 0.558 ; Not significant

VERBAL REASONING (V2)

GROUP DEMOGRAPHICS

NUMBER OF CASES : 464

EDUCATION : ‘O’ level, ‘A’ level, Graduate (mostly ‘A’ level)

GROUP DESCRIPTION : Factory workers, manufacturing applicants, engineering trades people, managers

AGE RANGE : 20 Years to 58 Years (Median age = 25 Years) (Mean age = 29.33 ; N = 332)

SEX BREAKDOWN : Female 77 (17.2%) Male 371 (82.8%)

TEST STATISTICS

TEST SCORE : Mean = 13.97 ; SD = 5.48, N = 464 ; Score Range : 0 to 30

TEST RELIABILITY : Cronbach’s Alpha = 0.8280

CORRELATIONS : AGE WITH TEST SCORE : -0.5569 (N = 332)
* Younger candidates score significantly higher
* Younger candidates have higher educational level

AGE DIFFERENCES : Up to and including 25 years : Mean = 17.8824 ; SD = 4.707 ; N = 170
26 Years upwards : Mean = 11.5679 ; SD = 4.753 ; N = 162
Mean Difference : 6.3145 ; 2-tail sign. = 0.000 ; Significant

SEX DIFFERENCES : Female : Mean = 18.4545 ; SD = 4.954 ; N = 77
Male : Mean = 13.1456 ; SD = 5.201 ; N = 371
Mean Difference : 5.3090 ; 2-tail sign. = 0.010 ; Significant

Applied Power (A1)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	384
EDUCATION :	GCSE, ‘O’ Level, ‘A’ Level (mostly ‘O’ level)
GROUP DESCRIPTION :	Clerical workers with large financial institutions
AGE RANGE :	17 Years to 37 Years (Median age = 22 Years) (Mean age = 22.26, N = 374)
SEX BREAKDOWN :	Female 71 (18.5%) Male 313 (81.5%)
TEST STATISTICS
TEST SCORE :	Mean = 11.82 ; SD = 4.39 ; N = 384 ; Score Range : 0 to 25
TEST RELIABILITY :	Cronbach’s Alpha = 0.8771
CORRELATIONS :	AGE WITH TEST SCORE : -0.0875 (N = 374) * Not Significant
AGE DIFFERENCES :	Up to and including 22 years : Mean = 12.2617 ; SD = 4.505 ; N = 214 23 Years upwards : Mean = 11.3313 ; SD = 4.297 ; N = 160 Mean Difference : 0.9304 ; 2-tail sign. = 0.040 ; Significant
SEX DIFFERENCES :	Female : Mean = 12.7887 ; SD = 4.724 ; N = 71 Male : Mean = 11.6038 ; SD = 4.292 ; N = 313 Mean Difference : 1.1849 ; 2-tail sign. = 0.040 ; Significant

PROCESSING SPEED (C1)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	555
EDUCATION :	GCSE, ‘O’ Level, ‘A’ Level (mostly ‘O’ Level)
GROUP DESCRIPTION :	Clerical workers with large financial institutions
AGE RANGE :	-
SEX BREAKDOWN :	Female 425 (81.1%) Male 99 (18.9%)
TEST STATISTICS
TEST SCORE :	Mean = 19.91 ; SD = 6.63 ; N = 555 ; Score Range : 0 to 42
TEST RELIABILITY :	Cronbach’s Alpha = 0.8906
CORRELATIONS :	AGE WITH TEST SCORE : -0.1647 (N = 225) * Younger candidates score significantly higher
AGE DIFFERENCES :	-
SEX DIFFERENCES :	Female : Mean = 20.5388 ; SD = 6.414 ; N = 425 Male : Mean = 17.3838 ; SD = 7.132 ; N = 99 Mean Difference : -3.1550 ; 2-tail sign. = 0.000; Significant

MECHANICAL REASONING (M1)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	718
EDUCATION :	No Formal Quals to Graduate (mostly ‘A’ level)
GROUP DESCRIPTION :	Engineers, Engineering Tradespeople, Factory workers, manufacturing applicants, pilot applicants
AGE RANGE :	17 Years to 60 Years (Median age = 23 Years) (Mean age = 26.83 ; N = 616)
SEX BREAKDOWN :	Female 49 (5.6%) Male 659 (94.4%)
TEST STATISTICS
TEST SCORE :	Mean = 15.09 ; SD = 5.78, N = 718 ; Score Range : 0 to 31
TEST RELIABILITY :	Cronbach’s Alpha = 0.7868
CORRELATIONS :	AGE WITH TEST SCORE : -0.2519 (N = 616) * Younger candidates score significantly higher
AGE DIFFERENCES :	Up to and including 23 years : Mean = 16.4343 ; SD = 5.450 ; N = 327 24 Years upwards : Mean = 14.4948 ; SD = 5.810 ; N = 289 Mean Difference : 1.9396 ; 2-tail sign. = 0.000 ; Significant
SEX DIFFERENCES :	Female : Mean = 12.1429 ; SD = 5.817 ; N = 49 Male : Mean = 15.3171 ; SD = 5.743 ; N = 659 Mean Difference : -3.1743 ; 2-tail sign. = 0.000 ; Significant

NUMERICAL COMPUTATION (N1)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	793
EDUCATION :	No Formal Quals to Graduate (mostly ‘A’ level)
GROUP DESCRIPTION :	Engineers, Engineering Tradespeople, Pilot applicants, Factory workers, Manufacturing applicants
AGE RANGE :	17 Years to 60 Years (Median age = 24 Years) (Mean age = 27.32 ; N = 669)
SEX BREAKDOWN :	Female 57 (4.4%) Male 724 (95.6%)
TEST STATISTICS
TEST SCORE :	Mean = 13.80 ; SD = 6.27, N = 793 ; Score Range : 0 to 37
TEST RELIABILITY :	Cronbach’s Alpha = 0.8613
CORRELATIONS :	AGE WITH TEST SCORE : -0.1655 (N = 142) * Younger candidates score significantly higher
AGE DIFFERENCES :	Up to and including 23 years : Mean = 14.6425 ; SD = 5.809 ; N = 328 24 Years upwards : Mean = 13.7462 ; SD = 6.479 ; N = 341 Mean Difference : 0.8963 ; 2-tail sign. = 0.060 ; Not significant
SEX DIFFERENCES :	Female : Mean = 13.0063 ; SD = 5.515 ; N = 57 Male : Mean = 13.8718 ; SD = 6.356 ; N = 724 Mean Difference : -0.8655 ; 2-tail sign. = 0.318; Not Significant

NUMERICAL REASONING (N2)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	368
EDUCATION :	From ‘O’ Level to Graduate (mostly ‘A’ level)
GROUP DESCRIPTION :	Engineers, Engineering tradesmen (fitters, electricians etc.), Managers, Pilot applicants, Factory workers, Manufacturing applicants
AGE RANGE :	17 Years to 54 Years (Median age = 23 Years) (Mean age = 25.38 ; N = 255)
SEX BREAKDOWN :	Female 90 (25.6%) Male 262 (74.4%)
TEST STATISTICS
TEST SCORE :	Mean = 10.89 ; SD = 4.98, N = 368 ; Score Range : 0 to 25
TEST RELIABILITY :	Cronbach’s Alpha = 0.8251
CORRELATIONS :	AGE WITH TEST SCORE : -0.3710 (N = 255) *Younger candidates scored significantly higher.
AGE DIFFERENCES :	Up to and including 22 years : Mean = 13.7869 ; SD = 4.484 ; N = 122 23 Years upwards : Mean = 10.7519 ; SD = 4.831 ; N = 133 Mean Difference : 3.0350 ; 2-tail sign. = 0.000 ; Significant
SEX DIFFERENCES :	Female : Mean = 11.8333 ; SD = 4.327 ; N = 90 Male : Mean = 10.7557 ; SD = 5.159 ; N = 262 Mean Difference : 1.0776 ; 2-tail sign. = 0.076 ; Not significant

PERCEPTUAL REASONING (P1)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	283
EDUCATION :	No formal quals to ‘A’ level (mostly ‘O’ level)
GROUP DESCRIPTION :	Factory workers, manufacturing applicants, engineering trades people
AGE RANGE :	17 Years to 54 Years (Median age = 22 Years) (Mean age = 26.71 ; N = 159)
SEX BREAKDOWN :	Female 20 (7.5%) Male 247 (92.5%)
TEST STATISTICS
TEST SCORE :	Mean = 11.32 ; SD = 3.65, N = 283 ; Score Range : 0 to 20
TEST RELIABILITY :	Cronbach’s Alpha = 0.7387
CORRELATIONS :	AGE WITH TEST SCORE : -0.2522 (N = 159) * Younger candidates score significantly higher
AGE DIFFERENCES :	Up to and including 21 years : Mean = 12.7805 ; SD = 3.410 ; N = 82 22 Years upwards : Mean = 11.2857 ; SD = 3.713 ; N = 77 Mean Difference : 1.4948 ; 2-tail sign. = 0.009 ; Significant
SEX DIFFERENCES :	Female : Mean = 12.8000 ; SD = 3.636 ; N = 20 Male : Mean = 11.2753 ; SD = 3.587 ; N = 247 Mean Difference : 1.5247 ; 2-tail sign. = 0.069 ; Not significant

SPATIAL ABILITY (S1)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	590
EDUCATION :	From ‘O’ level to Graduate (mostly ‘A’ level)
GROUP DESCRIPTION :	Engineers, Engineering Trades people (fitters, electricians), pilot applicants, creatives
AGE RANGE :	16 Years to 59 Years (Median age = 23 Years) (Mean age = 25.66 ; N = 495)
SEX BREAKDOWN :	Female 32 (5.4%) Male 558 (94.6%)
TEST STATISTICS
TEST SCORE :	Mean = 9.27 ; SD = 4.15, N = 590 ; Score Range : 0 to 21
TEST RELIABILITY :	Cronbach’s Alpha = 0.7891
CORRELATIONS :	AGE WITH TEST SCORE : -0.2878 (N = 495) * Younger candidates score significantly higher * Younger candidates have higher educational level
AGE DIFFERENCES :	Up to and including 23 years : Mean = 10.4148 ; SD = 3.719 ; N = 270 24 Years upwards : Mean = 8.1467 ; SD = 4.399 ; N = 225 Mean Difference : 2.2681 ; 2-tail sign. = 0.000 ; Significant
SEX DIFFERENCES :	Female : Mean = 10.1875 ; SD = 4.083 ; N = 32 Male : Mean = 9.2133 ; SD = 4.147 ; N = 558 Mean Difference : 0.9742 ; 2-tail sign. = 0.196 ; Not significant

VERBAL COMPREHENSION (V1)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	162
EDUCATION :	From ‘O’ level to Graduate (mostly ‘A’ level)
GROUP DESCRIPTION :	Managers, engineering tradespeople
AGE RANGE :	18 Years to 60 Years (Median age = 21 Years) (Mean age = 26.25 ; N = 134)
SEX BREAKDOWN :	Female 13 (8.7%) Male 137 (91.3%)
TEST STATISTICS
TEST SCORE :	Mean = 21.20 ; SD = 10.79, N = 162 ; Score Range : 0 to 57
TEST RELIABILITY :	Cronbach’s Alpha = 0.8380
CORRELATIONS :	AGE WITH TEST SCORE : 0.2273 (N = 134) * Older candidates score significantly higher
AGE DIFFERENCES :	Up to and including 21 years : Mean = 17.3919 ; SD = 6.257 ; N = 74 22 Years upwards : Mean = 24.0833 ; SD = 12.395 ; N = 60 Mean Difference : -6.6914 ; 2-tail sign. = 0.000 ; Significant
SEX DIFFERENCES :	Female : Mean = 22.1538 ; SD = 7.625 ; N = 13 Male : Mean = 20.3650 ; SD = 0.713 ; N = 137 Mean Difference : -1.7889 ; 2-tail sign. = 0.558 ; Not significant

VERBAL REASONING (V2)
GROUP DEMOGRAPHICS
NUMBER OF CASES :	464
EDUCATION :	‘O’ level, ‘A’ level, Graduate (mostly ‘A’ level)
GROUP DESCRIPTION :	Factory workers, manufacturing applicants, engineering trades people, managers
AGE RANGE :	20 Years to 58 Years (Median age = 25 Years) (Mean age = 29.33 ; N = 332)
SEX BREAKDOWN :	Female 77 (17.2%) Male 371 (82.8%)
TEST STATISTICS
TEST SCORE :	Mean = 13.97 ; SD = 5.48, N = 464 ; Score Range : 0 to 30
TEST RELIABILITY :	Cronbach’s Alpha = 0.8280
CORRELATIONS :	AGE WITH TEST SCORE : -0.5569 (N = 332) * Younger candidates score significantly higher * Younger candidates have higher educational level
AGE DIFFERENCES :	Up to and including 25 years : Mean = 17.8824 ; SD = 4.707 ; N = 170 26 Years upwards : Mean = 11.5679 ; SD = 4.753 ; N = 162 Mean Difference : 6.3145 ; 2-tail sign. = 0.000 ; Significant
SEX DIFFERENCES :	Female : Mean = 18.4545 ; SD = 4.954 ; N = 77 Male : Mean = 13.1456 ; SD = 5.201 ; N = 371 Mean Difference : 5.3090 ; 2-tail sign. = 0.010 ; Significant

References

Bartram D. 1992, private communication. Bartram reviewed the PPM in The Review of Psychometric Tests for Assessment in Vocational Training 1992. The Department of Employment, Great Britain.

Bennett Mechanical Comprehension Test 1980. Bennett G. The Psychological Corporation.

Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences, 2nd edition.

Cronbach L. J. (1990). Essentials of Psychological Testing, fifth edition. Harper Row, New York.

Differential Aptitude Test (DAT) 1986. Bennett G., Seashore H. and Wesman A. The Psychological Corporation.

Employee Aptitude Survey, Technical Report, 1980, F.L. and W.W. Ruch, Psychological Services Inc., Los Angeles.

NIIP Engineering selection test battery, 1955, “A Preliminary Notice”, J.R. Morrisby, Educational and Industrial Test Services.

	Population-based norms; associated with items
	Population-based norms; calculated
	Derived/calculated score
(x)	Dealing with missing values: (0) ignore them (1) use regression to estimate an answer (2) use the middle scale point as answer (3) use a defined score as answer