PCRI manual cover_070910_Layout 1 7/9/10 3:35 PM Page 1

Parent-Child
Relat ionship Inventory
( PCRI )
Manual
Anthony B. Gerard, Ph.D.

Additional copies of this manual (W-293B) may be purchased from WPS.

Please contact us at 800-648-8857, Fax 310-478-7838 or our Web site.
W-293B
01-02_v4_ 1-2.qxd 6/28/10 2:24 PM Page 1

1
INTRODUCTION

The Parent-Child Relationship Inventory (PCRI) reflect major features of parenting and the parent-child rela-
assesses parents’ attitudes toward parenting and toward their tionship. Consistent with the idea that parenting skills de-
children. The PCRI yields a quantified description of the fine a positive dimension, high scores on the PCRI scales
parent-child relationship that complements other assessment indicate good parenting skills and low scores indicate poor
procedures used in clinical evaluations of children and parenting skills.
families. Rather than replacing qualitative evaluation of
parent-child interactions, the PCRI helps to put qualitative PCRI Scales and Validity Indicators
impressions in perspective by making normative compar- The PCRI has seven content scales and two validity
isons possible. indicators. Each of the content scales explores a specific
Standardized on more than 1,100 parents across the aspect of the parent-child relationship. These scales were
United States, the PCRI identifies specific aspects of the developed using a combination of empirical and rational
parent-child relationship that may cause problems, as well approaches, as described in the section of chapter 4 entitled
as giving an overall picture of the quality of the relationship. “Item Analysis and Selection.” One of the two validity indi-
In an era when fathers are increasingly expected to take an cators gauges the client’s tendency to give socially desirable
active role in parenting, the PCRI explicitly measures the responses. The other validity indicator, which is based on
attitudes and behaviors of both mothers and fathers. It is agreement between answers on select pairs of items, measures
assumed that the PCRI will often be administered to cou- the tendency to give inconsistent responses. A listing of the
ples, and there are separate norms for mothers and fathers. items in the content scales and the Social Desirability scale
Recent increases in child custody litigation and appears in Appendix A.
divorce mediation have created a demand for sophisticated Content scales. Of the 73 items included in the con-
assessment of the relationships between parents and chil- tent scales, 26 are keyed positively and 47 are keyed nega-
dren, and the PCRI may prove especially useful in child cus- tively. If an item is positively keyed, a response of agree or
tody settings and in other institutions that specifically strongly agree increases the score for the scale on which that
address the needs of children. Used in conjunction with item appears; conversely, if an item is negatively keyed, a
interviews and other forms of clinical assessment, the PCRI response of disagree or strongly disagree increases the scale
can be an important element in the making of custody score. High scores indicate positive parenting characteristics.
recommendations and in evaluating the possibility that a The Parental Support scale (SUP), which has 9 items,
parent is abusive. assesses the level of emotional and social support a parent
receives.
General Description The Satisfaction With Parenting scale (SAT) consists
of 10 items measuring the amount of pleasure and fulfill-
The PCRI is a 78-item, self-report questionnaire that ment an individual derives from being a parent.
can be administered to either an individual or a group in The 14-item Involvement scale (INV) examines the
about 15 minutes. It has a fourth-grade reading level. level of a parent’s interaction with and knowledge of his or
The items were selected to measure a wide range of her child.
parenting dispositions and behaviors. Some of the items The Communication scale (COM) consists of 9
present general attitudes toward being a parent, and others positively keyed items that assess a parent’s perception of
are intended to elicit responses specific to a parent’s rela- how effectively he or she communicates with a child.
tionship with a particular child. All of the items have a The Limit Setting scale (LIM) contains 12 items, all
Likert-type, 4-point response format: strongly agree, agree, negatively keyed, that focus on a parent’s experience disci-
disagree, and strongly disagree. Rather than providing a sin- plining a child.
gle score representing an individual’s overall ability in and The 10-item Autonomy scale (AUT) assesses the
satisfaction with parenting, items are arranged in scales that ability of a parent to promote a child’s independence.

1
01-02_v4_ 1-2.qxd 6/28/10 2:24 PM Page 2

2 Parent-Child Relationship Inventory (PCRI)

The Role Orientation scale (ROL), comprising 9 difficulty between parents and children. By quantifying
items, examines parents’ attitudes about gender roles in aspects of parent-child interactions, the instrument makes it
parenting. possible to verify clinical hypotheses about individual and
Protocol validity. The PCRI has two validity indica- family disturbances against a background of objective data.
tors: Social Desirability (SOC) and Inconsistency (INC).
The Social Desirability indicator consists of five items Limitations
that are rarely endorsed in the positive direction. A low SOC
score suggests that the parent is giving distorted responses Assessment of parent-child relationships requires clin-
intended to portray the parent-child relationship in an ical sensitivity and a thorough knowledge of the research on
unrealistically positive light. parent-child interaction. Moreover, relationships between
The Inconsistency indicator comprises 10 pairs of parents and their children do not exist in a vacuum. They are
highly correlated items. The response to one item in the pair embedded in a matrix of family, cultural, and socioeco-
in effect predicts the response to the other item. High scores nomic factors, all of which influence parental attitudes and
on this indicator suggest inattentive or random responding. behaviors. Designed for use by individuals with a back-
ground in psychological assessment, the PCRI is intended
Applications to be one measure of the characteristics of the parent-child
relationship. Given the complexity of parent-child interac-
The PCRI is intended for use in a wide range of con- tions, the inventory must never be used in isolation as the
texts, including both clinical and research settings. Because sole basis for clinical diagnoses, treatment decisions, or
it is multidimensional, the PCRI identifies specific areas of custody recommendations.
08-19_v4_ 8-19.qxd 6/28/10 2:26 PM Page 17

4
DEVELOPMENT

Interest among mental health professionals in the Despite considerable interest in the MCRE, use of
assessment and treatment of dysfunctional parent-child rela- this instrument to address the major issues outlined above
tionships has grown dramatically in recent years. This has been constrained by two major factors. First, the origi-
increased concern is not surprising. Almost daily, newspapers nal standardization of the MCRE was based on a small
and television news programs report repeated incidents of sample of volunteer mothers in the Midwest. The extent to
physical or sexual child abuse. For each such highly publi- which the normative data generalizes appropriately to
cized case, many more are dealt with privately or go unre- other groups is unknown. Second, and perhaps more
ported. Improved assessment methods are needed to support importantly, the MCRE was originally conceptualized as a
research into the causes of child abuse, and to identify parents measure of maternal attitudes toward child rearing; both
and children who might benefit from specific interventions. the present normative data as well as many of the individual
Recent increases in child custody litigation and the items do not apply to fathers. The increasing recognition of
greater involvement of mental health professionals in vol- the father’s role in child rearing almost automatically calls
untary or court-mandated mediation have created a specific into question the usefulness of a parent attitudes question-
demand for more sophisticated procedures to assess parent- naire designed for mothers only (Cath & Ross, 1982;
child relationships. In their work within the legal system, Lamb, 1981).
psychologists and other professionals are often called upon Furthermore, there is a need for new parenting
to consult on decisions that can have enormous impact on questionnaires that are psychometrically stronger and more
people’s lives. They may, for example, be asked to make current than those of the past (Holden & Edwards, 1989).
recommendations about custody, foster care, visitation Many of the instruments developed for circumscribed
rights, and parental competence. In the absence of standard- purposes a generation ago are still in general use. In some
ized tests for assessing parental attitudes and behaviors, instances, instruments designed essentially for laboratory
those charged with making these decisions must rely solely use have been used as true assessment tools, with little or no
on their clinical skills or adapt existing instruments to normative data and limited validation. Because attitudes
purposes for which they were not intended. toward marriage and child-rearing have changed markedly
At present, there are no generally agreed upon in the past two decades, the content of these questionnaires
standards for evaluating parenting skills. To begin with, the is often out of date. Although the need for good tools to
subtle nuances of the parent-child relationship are difficult assess parent-child relationships is perhaps greater than
to measure. Many factors influence the quality of the rela- ever, few instruments actually meet present standards.
tionships within a family, and it is notoriously difficult to In light of the need to redesign the MCRE and in
identify the sources of family dysfunction. For example, it is response to the call for new instruments of this type, a
widely recognized that the causes of child abuse are multiple completely new inventory for assessing parental attitudes
and interactive (Spinetta & Rigler, 1972; Young, 1976). was developed. The development project had the following
One measure that has been used to assess parental atti- specific goals:
tudes is the Mother-Child Relationship Evaluation (MCRE). 1. To create an objective measure of parental
Originally published by Roth in 1961, the MCRE is a brief, attitudes that would be useful for both clinical
self-report instrument designed to assess both normal and and research applications
problematic aspects of parenting. Based on a set of con- 2. To develop items that are equally appropriate
structs initially presented by Symonds (1949), the MCRE for both males and females, and that are not
contains four subscales of 12 items each: Overprotection, age dependent
Overindulgence, Rejection, and Acceptance. Mothers are 3. To use items at a reading level that would
asked to rate their strength of agreement or disagreement make the inventory useful with low SES
with each item on a 5-point Likert scale. Clinical interpreta- populations
tion of the MCRE is based on individual item responses, the 4. To meet contemporary psychometric standards
clinical scale scores, the overall profile pattern, and the 5. To determine whether multiple scales would
integration of the test data with information from other facilitate accurate interpretation
sources, especially clinical interviews.

17
08-19_v4_ 8-19.qxd 6/28/10 2:26 PM Page 18

18 Parent-Child Relationship Inventory (PCRI)

Scale Construction Subjective Feedback

Important information on the wording and interpreta-
The first step in scale construction was to determine tion of each item came from individuals—some professionals,
the dimensions of parenting on which to generate items. some parents, some both professionals and parents—who
Whereas numerous, more specific content dimensions have carefully examined the 345-item Form B and commented on
been identified, factor analytic studies consistently extract all objectionable items. Information about the appropriateness
three superordinate dimensions of parenting attitudes: an of items also came unsolicited from many individuals who
affective dimension (“good/bad parenting”); a dimension completed the inventory after having been told that it was an
centering around control, authority, and limit setting; and a experimental instrument not suitable for use as a basis for
third dimension that varies more across studies but that important life decisions. Other individuals provided feed-
typically reflects either overprotection or punishment back by agreeing to “think aloud” as they went through each
(Whitman & Zachary, 1986). item, so that salient comments could be recorded by another
Factor analysis of data collected on a preliminary, individual. Observations about item clarity and acceptability
106-item version of the PCRI (Form A) identified five major gathered in these ways were used only at the end of the item
dimensions of parenting, and the items were organized into selection process to refine the wording of individual items.
scales based on these factors. New factor analyses were run
on each of these scales. The results suggested the organiza- Item Analysis and Selection
tion of the PCRI into 5 factor scales and 14 clinical scales. The final selection of items was based on a combina-
A set of 24 items for each scale, balanced for strength tion of empirical and rational criteria, following a systematic
and the direction of the response, was written for each of procedure originally presented by Jackson (1970). Much of
the 14 clinical scales, and a 9-item social desirability scale the data used in this step came from administering the
was added. The result was a 345-item version of the PCRI 345-item Form B to 211 parents living in and around
(Form B). Because this version contains many more items St. Louis, Missouri.
than is desirable for most practical applications, further item Step 1. The first step was to eliminate high- and low-
selection studies were required. frequency items—items everyone endorsed or no one
This research took three forms: (a) rating of items by endorsed. Such items contribute little unique information,
expert judges; (b) qualitative feedback from professionals and 91 items with extreme probabilities of endorsement or
and test takers to identify objectionable items; and rejection (above .90 or below .10) were discarded in this way.
(c) collection of additional empirical data and subsequent Step 2. The rank order of the items based on the
item analysis. ratings by the expert judges was examined for each scale,
and only items ranked 1 through 9 (out of 24) were retained.
Expert Rating Study These were all items that the expert judges perceived as
The 345-item version of the PCRI was rated by 11 well-written and meaningfully related to the content
judges chosen to represent a diversity of opinion. This group domains specified by the scale names. This procedure
included item-writing experts, clinicians and school eliminated an additional 61 items.
psychologists in active practice, a nationally known figure Step 3. An item that did not correlate significantly
on child abuse, and a minority psychologist interested (>.30) or uniquely with the total for its assigned scale was
in assessment. examined. Although the majority of such items were sub-
For the ratings, the individual items were typed onto sequently eliminated, attention was paid to the content of
3 × 5 index cards and presented in packs corresponding to the item. If the shift of an item to a new scale was defensible,
scales. The actual procedure was a Q sort (Nunnally, it was usually shifted. In part for structural reasons
1978). To insure useful response variance, the judges were connected with the original development of the scales
required to sort the items into five piles arranged in a trian- and in part to preserve face validity, an item that was
gular pattern approximating a normal distribution. Items more highly correlated with another scale than with its
were to be put in Pile 1 if they were extremely problematic own was sometimes retained on its original scale. An
or bore little relation to the intended content domain; item that had a low correlation with its own scale, how-
items were to be put in Pile 5 if they were superior and ever, and did not fit logically or statistically with another
should be retained; Piles 2, 3, and 4 represented the middle scale was eliminated.
ground. Systematic empirical criteria for item selection were
The average rating for each item across all 11 judges used, but they were not followed rigidly. The content of each
was computed, as was the rank order of the item within its item and its relationship to the other items in the scale were
particular scale. Because there were 24 items on each of the considered most important. The balancing of the best possi-
14 content scales, the rank orderings ranged from 1 to 24 ble psychometric properties against a subjective sense of
(where a ranking of 1 is highest); when there were ties, all which items logically belonged to which scale required
items having the same mean rating were given the same more than 20 iterations.
rank. These rank orderings were used later as part of the Step 4. Following Jackson (1970), a Differential
empirical criteria for item selection. Reliability Index (DRI) was computed for each item. This
08-19_v4_ 8-19.qxd 6/28/10 2:26 PM Page 19

Chapter 4 Development 19

procedure compares the corrected item-scale correlation to Table 1

the magnitude of the correlation between the item and the Scale Alphas for Development and
Social Desirability (SOC) scale. Because the SOC scale had Cross-Validation Samples
an alpha coefficient of .76, it was not ideal for this purpose
but did provide a systematic method for eliminating items Scale Development Cross-Validation
that were heavily biased with respect to social desirability.
Items with a DRI of less than .30 were usually eliminated. SUP .70 .71
Step 5. A major concern throughout the scale con- SAT .83 .85
struction and item analysis process was that the final scale INV .77 .77
be perceived as logical and useful by clinicians, school psy- COM .81 .82
chologists, and others who would use the scale for various LIM .88 .87
applications. Achieving this goal implied a balance between AUT .80 .78
the ideals of empirical test construction and the demands of ROL .77 .75
clinical usefulness. The final step in the initial development
Note. The samples for Development (n = 519) and Cross-Validation
procedure was to give the revised scales to a new set of five (n = 518) were derived from the normative sample.
expert judges. Rather than assigning each item a value from
1 to 5, this panel of judges made dichotomous decisions
about whether a particular item was good or bad and Social Desirability
whether it belonged on a particular scale. Based on this The purpose of a social desirability scale is to identify
input, items were eliminated because the judges found them clients who are responding with a so-called “defensive
objectionable or because they could not be assigned sensibly response set.” On a test like the PCRI, individuals may have
to a particular scale. Although a few problematic items iden- many reasons to “fake good.” For example, a client may be
tified at this stage were reassigned to other scales, most were involved in a custody dispute, which places a heavy premium
simply eliminated. on appearing to be an excellent parent. Even when less is at
With these adjustments, the PCRI consisted of 107 stake, many individuals tend to present themselves in a
items distributed over two validity scales and eight content favorable light. Signs that a client is “faking good” intro-
scales. This version of the instrument was used over a period duce a note of caution into the interpretation of his or her
of approximately 4 years to conduct a variety of validity scores, and may be sufficient reason to treat the results as
studies. The 107-item form was also used to collect the entirely invalid.
standardization data. The present five-item Social Desirability scale is the
Step 6. Following the collection of the standardization result of the same item selection procedure described in the
data an additional item selection process was initiated, section of this chapter entitled “Scale Construction.” All of
similar to that described in Step 3. Scale alpha levels and the items were written to reflect the exaggeration of positive
item-total correlations were obtained using approximately qualities characteristic of a defensive response style (faking
half of the normative sample (n = 519). Items that were not good). The mean score on the SOC scale was 14.52, and the
contributing to scale reliabilities were eliminated. In a few standard deviation was 2.63. The items on this scale and the
cases items were shifted from one scale to another, based on endorsement percentage for each are presented in Table 2.
a combination of empirical and rational criteria. One scale, The cutting score of 9 on the SOC scale is based on
Moderation, was eliminated, and two of its items were simple inspection of the responses of those in the standard-
incorporated into the Autonomy scale. The results of this ization sample. Those with scores of 9 or lower constitute
procedure were cross-validated using the remaining half of 5% of the sample. To use this proportion to identify those
the normative sample (n = 518). The scale reliabilities who may be faking good appears to be a sufficiently selec-
obtained using both samples are presented in Table 1. As the tive screening.
values in the table show, shrinkage in the scale reliabilities was
minimal, supporting the validity of the new scale structure. Inconsistency
The result of these additional adjustments was a The purpose of the Inconsistency indicator is to
78-item instrument with a fourth-grade Flesch-Kincaid identify protocols that result from inattentive responding.
reading level (Thomas, Hartley, & Kincaid, 1975). Almost by definition, such a response set will produce
patterns that resemble random responding. Although
Development of Validity Indicators random responding is obviously undesirable, it is also
relatively easy to detect with some certainty. The chief prob-
The two response validity indicators for the PCRI lem in designing the INC indicator is to arrive at a criterion
were generated in different ways. The Social Desirability for identifying a protocol as invalid.
scale was developed at the same time as the content scales. The INC indicator is based on interitem contingen-
The Inconsistency indicator was developed using the results cies. Ten pairs of highly correlated items were identified.
of the standardization study. The cut points for both scales Each pair consists of items from the same scale; intrapair
were established using standardization results. correlations range from .53 to .74, and all of the correlations
20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 20

20 Parent-Child Relationship Inventory (PCRI)

Table 2
Percent Endorsement of PCRI Social Desirability Scale Items

Item Percent Strongly Agree

18. My child is never jealous of others. 6.4

30. I never worry about my child. 1.6
37. I have never had any problems with my child. 4.5
43. I have never been embarrassed by anything my child has said or done. 8.1
47. My child never puts off doing things that should be done right away. 1.7

Note. N = 1,093.

are positive. Scores equal the number of pairs in which the oped by Cureton (1957), Dawes (1962), and Dawes and
items are endorsed in opposite directions, with the limit that Meehl (1966), and applied by Cull and Gill (1988), among
there must be a 2-point difference between the ratings of the others, it is assumed that the distribution of “good” proto-
items in a pair. The item pairs comprised by the INC indica- cols numerically (or spatially) overlaps the distribution of
tor are as follows: 17 and 36, 24 and 27, 55 and 67, 64 and “bad” protocols. For each possible INC score (1 through 10)
77, 53 and 60, 63 and 72, 39 and 46, 10 and 26, 49 and 78, a thin statistical “slice” is taken through both distributions,
and 25 and 32. The mean score was .42, and the standard permitting an estimate of the proportions of good protocols
deviation was .77. Inconsistent responses to two or more of and bad protocols that attain that score. The deeper into the
these pairs were rare in the normative sample, occurring in bad distribution, the higher the proportion of bad protocols a
fewer than 6% of the cases. Consequently, the validity of a given score represents.
protocol is questioned if it contains inconsistent responses to Using this procedure, it is found that the exact proba-
more than one of these pairs of items. bility that an INC score of 1 comes from a sample of bad
The validity of using a cut point of one pair on the (random) protocols is 29.1%, but the probability that an INC
INC indicator was subjected to statistical test. It is possible score of 2 comes from a sample of bad protocols is 84.4%.
to compare the probability that a given score on the INC For scores above 3, it is a virtual certainty that the protocol
indicator comes from a sample of good protocols (those that is bad. The large classification increment between 1 and 2
result from attentive responding) with the probability that justifies the establishment of 1 as the INC cut point. There is
the same score comes from a sample of bad protocols (those a high likelihood that protocols with INC scores greater than
resulting from random responding). Following logic devel- 1 reflect inattentive or random responding.
20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 21

5
STANDARDIZATION

This chapter describes the methods employed in deriv- filled out. In some instances, however, the packets were
ing norms for the PCRI. The first section describes the data distributed and completed as part of a group administration
collection effort, including the sample selection procedures. at a PTA meeting or other parents’ gathering.
Subsequent sections discuss the reduction and analysis of the Each site received detailed data collection instruc-
normative data, as well as the structure of the PCRI norms. tions. The individual packets also contained detailed
instructions to the parents for taking the test and filling out
Standardization Sample the accompanying demographic form. None of the sites
reported any difficulty in following the instructions or
The normative data for the PCRI represents the administering the instrument.
responses of more than 1,100 mothers and fathers. Data
was collected through schools and day-care centers in each Sample Characteristics
of the four major geographical regions of the United States The main sample demographics are given in Table 3.
(Northeast, South, Midwest, and West). The first of the Where it is appropriate, Table 3 also gives U.S. Census
following sections describes the procedures for acquiring percentages corresponding to sample figures. In general, the
standardization sites and collecting data; the second sample was geographically diverse but weighted heavily
section discusses the demographic characteristics of the toward the middle of the socioeconomic spectrum. Sample
normative sample. characteristics are described briefly in the following sections.
Age and sex. That the PCRI is a self-report of parents
Data Collection about their relationship with their children slightly compli-
The collection sites for the PCRI normative data were cates the collection and description of the normative data.
schools and day-care centers. The possibility that parents of children at different ages
Letters soliciting participation in the study were sent respond differently to their children meant that it was neces-
to 2,000 individuals identified as principals of elementary sary to make the age of the child a factor in the normative
schools, middle schools, and junior high schools or as direc- study. In fact, age of child was presupposed to be a factor of
tors of preschools and day-care centers. Those interested in greater interest than age of parent. Consequently, data was
participating were to return a postage-paid card requesting collected through institutions expected to have students as
basic demographic information about the parents in their young as 3 years of age and at institutions expected to have
institutions, and 88 such cards were returned. The light students as old as 15 years of age. The sex of the child was
return rate (4.4%) was anticipated, and the number of also expected to exert some influence on PCRI scores. For
returns was in excess of that needed to fulfill the require- that reason, sex of child was also a factor in the normative
ments of the normative study. study. The distribution across sex and ages of children and
Based on geographic distribution and on the demo- parents is given in Table 3.
graphics supplied on the return cards, 18 data collection Responses of mothers and fathers. Parenting instru-
sites were selected. Each site received a specified number of ments have typically been based on the questionnaire
data collection packets, and each packet contained enough responses of mothers (Holden & Edwards, 1989). As dis-
materials so that both parents of a given child could partici- cussed in the introductory section of chapter 1, a main moti-
pate in the study. In most cases it was the joint responsibility vation for developing the PCRI was to present normative
of the parents’ organization at the school and a designated information on the attitudes of both parents toward children
member of the school administration to distribute the pack- and child-rearing. Although there are more mothers than
ets, see that they were properly completed, collect them, and fathers in the sample, the sites participating in the standard-
return them for analysis. Also in most cases, the packets ization of the PCRI were, by and large, successful in collect-
were sent home with students and returned after they were ing responses from fathers, as the figures in Table 3 suggest.

21
20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 22

22 Parent-Child Relationship Inventory (PCRI)

Table 3
Sample Demographics

Child’s Age <3 3 4 5 6 7 8 9 10 11 12 13 >13

n 54 56 122 181 145 121 89 129 105 68 43 42 37
Sample % 4.5 4.7 10.2 15.2 12.2 10.2 7.5 10.8 8.8 5.7 3.6 3.5 3.2

Child’s Sex Male Female

n 595 588
Sample % 50.2 49.7
Censusa 51.2 48.8

Relationship to Child Mother Father Stepparent Other

n 669 474 33 37
Sample % 55.2 39.1 2.7 3.1

Parents’ Age 18–24 25–34 35–44 45–54 >55

n 27 389 657 95 17
Sample % 2.3 32.8 55.4 8.0 1.4
Censusb 13.8 22.9 19.1 13.0 27.5

Native
Ethnic Backgroundc Asian Black Hispanic American White Other
n 22 81 18 12 1,018 35
Sample % 1.9 6.9 1.5 1.0 85.7 3.0
Censusd 2.9 12.1 9.0 0.8 80.3 —

Parents’ Less Than Four Years

Education High School High School Some of College
Level Graduation Graduation College or More
n 52 381 322 413
Sample % 4.5 32.6 27.6 35.4
Censuse 13.5 40.0 21.1 25.4

Region Northeast South Midwest West

n 338 491 201 147
Sample % 28.7 41.7 17.1 12.5
Censusd 20.4 34.4 24.0 21.2

Note. N = 1,192; sample size varies slightly across subtables.

a
Based on U.S. Census figures for children (U.S. Bureau of the Census, 1991).
b
Based on U.S. Census figures for those over 16 (U.S. Bureau of the Census, 1991).
c
Based on respondents’ reports of children’s ethnicity.
d
Based on U.S. Census figures (U.S. Bureau of the Census, 1991).
e
Based on U.S. Census figures for adults 25–44 (U.S. Bureau of the Census, 1991).

Socioeconomic status (SES) and ethnicity. The their children’s ethnicity was used as the primary indicator
figures in Table 3 also show that the PCRI standardization of ethnicity. The actual difference between respondents’
sample is somewhat better educated and less diverse than reports of their own ethnicity and their children’s was small.
the U.S. population as a whole. For example, the median As shown in Table 3, whites are overrepresented in the
number of years of education for the sample was 14, whereas normative sample.
the median for U.S. residents between 25 and 44 (a span that Region. Of the 18 schools that agreed to participate,
encompasses virtually all of the participants in the norma- 13 returned data—4 in the South and 3 each in the
tive study) is approximately 13 (U.S. Bureau of the Census, Northeast, Midwest, and West. In each region, one of the
1991). The figures for parents’ occupational status, which institutions returning data was a preschool or day-care
were also collected, closely parallel those for education; center. Given that the South has a substantially larger popu-
approximately 54% of the sample participants were lation than do the other major geographic regions (U.S.
employed in technical, managerial, or professional positions. Bureau of the Census, 1991), the number of participants in
Because the bulk of the data comes from both parents’ each region, presented in Table 3, is roughly proportional to
ratings of a single child, the respondents’ identification of the distribution of population across regions.
20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 23

Chapter 5 Standardization 23

Effects of Sample Characteristics of spuriously significant findings posed little danger, and it
was actually conservative not to control for the acceptance
Because divergence of standardization sample charac- of spurious effects through the use of a technique for adjust-
teristics from population characteristics could produce ing the size of the statistical rejection region.
anomalies in the PCRI norms, sample characteristics were Effect size. It is necessary to place those statistically
examined for their influence on PCRI scores. The effects of significant results that do arise in their proper perspective.
sample characteristics were evaluated for their statistical Accordingly, the standard error of measurement (SEM) was
significance through the use of multivariate analysis of vari- chosen as the criterion for assessing the practical importance
ance (MANOVA), and the statistical results were evaluated of statistically significant effects. The SEM, which corre-
for their practical importance through the use of standard sponds to the 68% confidence interval, is frequently used in
techniques for assessing the meaning of individual test test interpretation. It therefore provides a familiar bench-
results. Simplifying assumptions were made in order to mark for assessing the meaning of effects.
reduce the complexity of the analysis, and criteria were On each scale, differences among group means and
developed for assessing the relationship of statistically differences of group means from the overall mean have been
significant group effects to the interpretation of individual interpreted through the application of the SEM to individual
PCRI scores. The effects of purely demographic factors scores. Differences between means of greater than one SEM
(ethnicity, SES, and region) are discussed separately from are assumed to be of substantive importance to the clinical
the effects of factors more directly related to the content of interpretation of individual scores. No clinical importance is
the PCRI (age and sex of parent and child). attached to differences of less than one SEM .
The formula for the SEM is:
Evaluating Effects SEM = SD * 1 – r
Such an examination of the data set raises two
√
In this instance, the SD is always 10, because the scores in
methodological issues. First, the number of variables makes question are T-scores. In general, the r on the right side of
the number of potential comparisons so large that it would this equation is an estimate of the reliability of the scale in
be virtually impossible to determine the actual significance question; in this instance, the internal consistency reliability
of statistically rare results. Second, statistically significant of each scale (coefficient alpha) is used as the value of r.
group differences may or may not reflect differences of (The scale alphas for the PCRI are discussed in chapter 6 in
practical significance in evaluating individual PCRI results. the section entitled “Internal Consistency.”) The interpreta-
Efforts were made to address both of these problems and to tion criteria arrived at in this way for the scales of the PCRI
render results comparable across subscales. are presented in Table 4. The reported values are based on
Standard metric. To assist in evaluating differences the preliminary T-scores described earlier, and they are
among groups on the seven content scales of the PCRI, all of rounded for simplicity.
the raw data was transformed to T-scores based on the mean
and standard deviation (SD) of each scale for the entire stan- Demographic Variables
dardization sample. The overall mean of the sample is set at Several analyses were done in order to evaluate the
50 and the standard deviation is set at 10. Therefore, the effects of purely demographic factors, namely ethnicity,
group mean for a scale is below 50 when the group has SES, and region. In some cases these analyses were run on
endorsed fewer than the average number of items in the stratified subsamples drawn from the standardization
scale-positive direction, and it is greater than 50 when the
group has endorsed greater than the average number of
items. This linear T-score transformation is not the T-score Table 4
transformation ultimately used in constructing the PCRI Standard Errors of Measurement (SEM s)
norms. Instead, it is a preliminary centering that represents and Criteria for Interpretation of Substantive
all of the scale results in terms of a common, familiar metric. Differences in the PCRI Normative Sample
The T-score transformation does not affect the presence and
size of effects in the data set. Interpretation
Effect significance. Two features of the analysis PCRI Scale SEMa Criterion
reduced its complexity and protected against mistakenly
SUP 5.42 5
identifying effects as significant (Type I error). First,
SAT 3.84 4
although the initial steps of the analysis included the assess-
INV 4.76 5
ment of interaction effects, no interaction terms were antici-
COM 4.24 4
pated, permitting the analysis to focus on simple main
LIM 3.49 3
effects; the MANOVA models employed were designed to
AUT 4.53 5
establish the need to pursue more detailed analysis of inter-
ROL 5.03 5
action effects. Second, the real purpose of the analysis was
to uncover and explain any statistically significant effects a
Computed using Cronbach’s alpha as the estimate of scale reliability and
that might distort the norms. Consequently, the acceptance based on preliminary T-scores for each scale.
20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 24

24 Parent-Child Relationship Inventory (PCRI)

sample. Each subsample was designed to approximate U.S. the mean Autonomy score for blacks was approximately 4.5
population proportions of the variables in question. T-score points below that of whites, almost precisely the
MANOVAs on these samples address the possibility that magnitude of the SEM for the autonomy scale.
interactions among demographic variables could produce There were significant differences related to level of
substantial distortions in the data. The analyses also reveal education on the support and autonomy scales. As might be
the effects of demographic variables on PCRI results. expected, parents with four or more years of college had a
Effects of ethnicity and education. A sample of 240 higher mean score on the support scale than did parents at
individuals stratified by education level and ethnicity to the other three levels of education, a difference that exceeds
approximate U.S. population proportions was drawn from the criterion for interpretation of a substantive difference.
the larger standardization sample. Because representation of Furthermore, the difference between the mean for parents
other ethnic groups in the sample was too small for a mean- with at least a college education and the overall mean
ingful analysis, only blacks and whites were included. A approximately equalled the criterion for interpretation of a
MANOVA was run with ethnicity of child and respondent substantive difference. Scores on the autonomy scale for
education treated as independent variables and scores on the parents with a high school education or less were lower than
seven content scales of the PCRI treated as dependent vari- the overall mean and lower than the scores of parents with at
ables. The interaction term was not significant, but there least some college by amounts that exceed the criterion for
were significant main effects of both independent variables. interpretation of a substantive difference.
For each of the independent variables, there were significant Regional differences. A one-way MANOVA was run
differences between groups on two of the seven PCRI with region as the independent variable and scores on the
scales, as noted in Table 5. The relevant cell means are seven content scales of the PCRI as dependent variables.
presented in the table where there are significant group (The relatively representative distribution of the sample
differences on a given scale. across regions precluded the need for the drawing of a
There were significant differences between blacks and stratified subsample.) There were significant regional
whites in their level of parental satisfaction and in their level differences on four of the seven scales, as shown in Table 6,
of autonomy, as measured with the PCRI. In both instances which includes the relevant weighted cell means for those
these differences appear to be substantive: The mean four scales. In no instance do differences between regions
Satisfaction With Parenting score of blacks was nearly 6 or differences of regions from the mean exceed the criterion
T-score points lower than that of whites, clearly exceeding for interpretation of a substantive difference.
the criterion for interpretation of a substantive difference;

Table 5
Analysis of Variance Results and Associated Group Means
for Ethnicity and Education

F Group Means

Ethnicity Black White

SUP .02 — —
SAT 7.72** 44.90 50.62
INV 1.53 — —
COM 2.08 — —
LIM .03 — —
AUT 5.15* 44.69 49.21
ROL .21 — —

Less Than Four Years

High School High School Some of College
Education Graduation Graduation College or More
SUP 2.88* 48.29 48.45 48.34 54.80
SAT 2.53 — — — —
INV 1.40 — — — —
COM 2.60 — — — —
LIM 2.56 — — — —
AUT 4.75** 42.39 44.82 51.66 52.25
ROL 1.52 — — — —

*p < .05; **p < .01.

20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 25

Chapter 5 Standardization 25

Content Variables seven scales, as shown in Table 10. As discussed in the

On a priori grounds, four variables are assumed to following section, the observed age differences were not
have the potential of interacting in important ways with used in the construction of norms for the PCRI. The group
factors the PCRI measures directly. They are the age of the differences shown in Table 10 may be used as a guide by
parent, the relationship of the parent to the child (the sex of those interested in adjusting the interpretation of individual
the parent), the sex of the child, and the age of the child. scores based on the age of the child.
Strong, interpretable relationships between these presum-
ably content-related variables and scores on the PCRI Effect Size and Norms Construction
content scales could have direct implications for the struc- With one exception, the differences reported here
ture of the PCRI norms. were not used as the basis for the construction of group-
Parent’s age. A one-way MANOVA was run with specific norms. In general, these differences were neither
parent’s age (four age levels) as the independent variable large enough nor pervasive enough to warrant the loss of
and scores on the seven PCRI content scales as dependent statistical power and convenience associated with the
variables. The small number of parents in the sample over construction of a series of separate norms tables.
the age of 54 was excluded from the analysis, as was the The pattern of results for ethnicity does not justify
small number of stepparents. There was a main effect for separate norms. The reliable difference on the Satisfaction
age. The results are presented in Table 7. As before, With Parenting scale does not, by itself, constitute evidence
relevant cell means are given where there are significant of systematic differences in parenting between blacks and
differences. whites. Nevertheless, the information presented in Table 5
Direct examination of the means for parents of does permit the making of slight adjustments in the interpre-
different ages suggests that there is a marked difference in tation of the PCRI profiles of black parents.
both range and magnitude between the mean PCRI scale From a wider perspective, it seems appropriate to
scores of parents 24 years old and younger and those of perceive the lower scores of low-SES individuals as
parents from 25 to 54. In part, this difference may be a possibly reflecting systematic disadvantages. The
result of the smaller sample of younger parents. When the observed differences may reflect a real need for some type
youngest parents are excluded from the analysis, there is of intervention. Norms that adjusted for differences based
no significant multivariate main effect of parent’s age nor on SES would tend to reduce the proportion of clinically
are there significant main effects for any of the PCRI significant elevations identified among low-SES indi-
content scales. viduals. Furthermore, the observed difference on the
Parent’s relationship to child. A one-way Parental Support scale transparently reflects the more
MANOVA was run with parent’s relationship to the child privileged high-SES circumstances, leaving the Autonomy
(mother or father) as the independent variable and scores on scale alone as a possible reflection of structural differ-
the seven PCRI content scales as the dependent variables. ences in parenting related to SES. Therefore, the norms for
As shown in Table 8, differences between mothers and the PCRI were constructed without specific adjustments
fathers are among the most pervasive of the effects exam- for SES.
ined here. Although the difference on no subscale exceeds Parent’s age was also not used as a basis for norms
the criterion for interpretation of a substantive effect, the construction. The sample of young parents was small. As
overall pattern of results strongly suggests a systematic with SES, the use of norms based solely on this segment of
difference between mothers and fathers in their responses on the sample might under-identify the presence of serious
the PCRI. Consequently, there will be separate sets of PCRI problems. The presentation of separate norms for these
norms for mothers and fathers, as described in a later section groups of special concern might obscure problems where
of this chapter (“Construction of the PCRI Norms”). they do exist.
Child’s sex. A one-way MANOVA was run with sex Of the variables closely related to the content of the
of child as the independent variable and scores on the seven PCRI, only the parent’s relationship to the child is a factor in
content scales of the PCRI as dependent variables. As the norms. Although the effects of the child’s age were even
shown in Table 9, there was a significant difference favoring more pervasive than were those of relationship, age was not
females only for the Satisfaction with Parenting scale, a used as the basis of norms construction for two reasons.
difference not exceeding the criterion for interpretation. First, the effects of age were pervasive, but they were based
Therefore, sex of child was not a factor in the construction on slightly smaller differences overall than were those for
of the PCRI norms. relationship. Second, the interests of statistical consistency
Child’s age. A one-way MANOVA was run with age compete with those of economy of presentation. The small
of child as the independent variable and scores on the seven but reliable differences accounted for by age of child do not
content scales of the PCRI as dependent variables. appear to justify complicating the presentation of the norms
(Respondent’s children were classified into three age levels: literally by a factor of three. In contrast, the claim of
5 years and younger; 6 through 10 years; 11 years and relationship (mother or father) as a basis for a distinction in
older.) There was a main effect for age of child; there were the norms is strong.
significant differences between age groups on six of the
20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 26

26 Parent-Child Relationship Inventory (PCRI)

Table 6
Analysis of Variance Results and Associated Group Means
for Region

F Group Means

Region Northeast South Midwest West

SUP 5.91*** 50.92 49.92 47.76 51.67
SAT 4.73** 51.20 49.76 49.04 50.61
INV 2.13 — — — —
COM .38 — — — —
LIM 2.68* 50.07 50.61 48.72 49.87
AUT 1.14 — — — —
ROL 5.60*** 49.50 50.95 51.02 46.83

*p < .05; **p < .01; ***p < .001.

Table 7
Analysis of Variance Results and Associated Group Means
for Parent’s Age

F Group Means

Parent’s Age 18–24 25–34 35–44 45–54

SUP 2.39 — — — —
SAT 10.46*** 38.76 50.68 50.84 49.73
INV 4.94** 42.94 51.24 50.47 49.59
COM 2.09 — — — —
LIM 1.20 — — — —
AUT 3.32 43.61 50.29 50.46 50.13
ROL .45 — — — —

**p < .01; ***p < .001.

Table 8
Analysis of Variance Results and Associated Group Means
for Relationship to Child

F Group Means

Relationship Mother Father

SUP 9.80** 48.83 51.90
SAT .28 — —
INV 14.09*** 52.24 48.01
COM 4.95* 51.76 48.25
LIM 8.74** 49.51 51.19
AUT .10 — —
ROL 4.57* 50.81 49.92

*p < .05; **p < .01; ***p < .001.

20-28_v4_20-27.qxd 6/28/10 2:28 PM Page 27

Chapter 5 Standardization 27

Table 9
Analysis of Variance Results and Associated Group Means
for Sex of Child

F Group Means

Sex Male Female

SUP .00 — —
SAT 4.40* 49.61 50.83
INV .02 — —
COM 1.03 — —
LIM 3.26 — —
AUT 2.06 — —
ROL .01 — —

*p < .05.

Table 10
Analysis of Variance Results and Associated Group Means
for Child’s Age

F Group Means

Child’s Age 5 and under 6–10 11 and over

SUP 5.07** 49.68 51.18 48.86
SAT 13.31*** 49.74 51.84 48.08
INV 7.08*** 49.98 51.31 48.34
COM 3.10* 50.17 50.83 48.73
LIM 4.64** 49.12 51.04 50.00
AUT 2.66 — — —
ROL 3.18* 49.12 50.80 50.12

*p < .05; **p < .01; ***p < .001.

Construction of the PCRI Norms scales was skewed, the T-score distributions for all of the
scales have been normalized. The normalization function
The final normative sample included 668 mothers and transforms the raw score distribution to fit a normal curve.
471 fathers, 1,139 parents in all. As discussed in the imme- Therefore, the T-scores are based on the cumulative proba-
diately preceding sections, respondents over 54 years of age bilities in the raw score distribution rather than on the mean
were excluded from the normative sample. (Protocols with and standard deviation of the scores directly (Anastasi,
more than 10% of the responses missing had already been 1988; Guilford & Fruchter, 1978).
excluded.) As an aid in interpreting scores, percentile equivalents
Separate norms tables for mothers and fathers are of the T-scores are given for the 99th, 95th, 90th, 75th, 50th,
presented in Appendix B. Scale scores are presented as 25th, 10th, 5th, and 1st percentiles.
T-scores with a mean of 50 and a standard deviation of 10. A The use of the norms in scoring is discussed in chapter
T-score conversion transforms raw scores so that the num- 2 of this manual in the section entitled “Scaling Scores and
ber of items on a scale does not greatly influence profile Displaying Results.”
elevations. Because the distribution of scores on some of the