A 6-item self-report questionnaire assessing children’s dispositional hope. The measure is “based on the premise that children are goal directed and that their goal-related thoughts can be understood according to two components: agency and pathways” (Snyder et al., 1997, p. 400). These two components, agency (ability to initiate and sustain action towards goals) and pathways (capacity to find a means to carry out goals), are assessed by the measure.
Overview
CHS
Snyder, C. R., Hoza, B., Pelham, W. E., Rapoff, J., Ware, L., Danovsky, M., Highberger, L., Rubinstein, H., & Stahl, K. (1997). The development and validation of the children’s hope scale. Journal of Pediatric Psychology, 22, 399-421.
The measure is available in the article cited above.
Administration
6-point Likert scale (1=None of the time, 2=A little of the time, 3=Some of the time, 4= A lot of the time, 5=Most of the time, 6=All of the time)
| Domains | Scale | Sample Items |
|---|---|---|
| Total Score | Agency | I think I am doing pretty well. |
| Pathways | I can think of many ways to get the things in life that are most important to me. |
Training
No information was found regarding training to administer this measure
Parallel or Alternate Forms
1. As part of the validity study, the authors developed a modified parent-report version of the Children's Hope Scale. Pronouns were changed from the first to the third person.
2. The measure is also sometimes referred to as the Children's Trait Hope Scale.
3. There are two related adult measures: the State Hope Scale and the Trait Hope Scale.
Psychometrics
| Type: | Rating | Statistics | Min | Max | Avg |
|---|---|---|---|---|---|
| Test-Retest | Acceptable | Pearson Correlation | 0.73 | 0.73 | 0.73 |
| Internal Consistency | Acceptable | Chronbach's alpha | 0.72 | 0.86 | 0.77 |
| Inter-rater | |||||
| Parallel/Alternate Forms |
TEST-RESTEST RELIABILITY
1 week test-retest reliability for the total score, assessed with 89 children, was r=.73, p<.001 (Snyder et al., 1997). This number is reported in the table, as it is most similar to the timeframe used for other measures.
Test-retest stability over a one month-interval assessed with 359 children was r=.71, p<.001 (Snyder et al., 1997).
INTERNAL CONSISTENCY
For internal consistency, Snyder et al. (1997) reported data only for the total score across 6 samples. The median alpha was .77, range: .72-.86. These data are presented in the above table.
For children aged 15-19 Valle et al. (2004) report an alpha for the total CHS as .84, with item-total correlations ranging from .51 to .69.
For children aged 10-14, Valle et al. (2004) reported an alpha for CHS total as .83, with item total correlations ranging from .55 to .68.
OTHER STUDIES
Ey, S., Hadley, W., Allen, D. N., Palmer, S., Klosky, J., Deptula, D., et al. (2005) reported alpha=.84 for the total scale in a sample of 204 3rd-6th graders.
From Snyder et al. (1997):
A set of 12 items measuring agency and pathway thinking were derived by the senior author’s research group. Agency items were related to an active “doing” orientation about the present and future. Pathways assessed how children reach goals under both ordinary and more stressful circumstances.
Items were piloted with children aged 8-16 to determine how well children understood the items. They were revised based on feedback to simplify reading level and increase understandability. The original 12-item version was then piloted with 197 boys and 175 girls aged 9-14. Principal components factor analysis with a varimax rotation was conducted to examine a 2-factor solution. Based on these data, 3 agency items and 3 pathways items that had weak or equivocal loading were discarded.
Factor analysis with the remaining 6 items resulted in two factors, the first accounting for 32.5% of the variance and the second accounting for 25.9% of the variance. To test the factor structure, additional data were gathered with the same children and were factor analyzed. The analysis resulted in two correlated factors (r=.61).
| Validity Type | Not known | Not found | Nonclincal Samples | Clinical Samples | Diverse Samples |
|---|---|---|---|---|---|
| Convergent/Concurrent | Yes | Yes | Yes | ||
| Discriminant | Yes | Yes | Yes | ||
| Sensitive to Change | |||||
| Intervention Effects | |||||
| Longitudinal/Maturation Effects | |||||
| Sensitive to Theoretically Distinct Groups | |||||
| Factorial Validity | Yes | Yes | Yes |
CORRELATIONS WITH OTHER MEASURES
1. Snyder et al. (1997): The CHS was found to correlate significantly with a modified parent-report version of the Children’s Hope Scale across different samples (r>.36 for all samples).
In addition, across samples, CHS scores were found to correlate significantlyand in the expected direction with scores on the five Subscales and the Global Self-Worth scale of the Self-Perception Profile for Children, the Children’s Perceived Physical Efficacy Scale, the Children’s Attributional Style
Questionnaire, and the Children’s Depression Inventory. The measure was also found to be related to measures of social desirability, which the authors interpreted as adaptive coping.
The measures included the Children’s Social Desirability Questionnaire and the Lie Scale of the Revised Children’s Manifest Anxiety Scale.
2. Valle, Huebner, & Suldo (2004) examined the psychometrics of the CHS with a sample of 460 students aged 15-19 from two public high schools.
Ethnicity of students was 52% African American, 40% Caucasian, 2% Asian American, 1% Hispanic, 1% Native American, and 4% other ethnic background. A majority were of lower socioeconomic status, with 63% reportedly qualifying
for a free or reduced lunch.
They reported significant correlations between the CHS and scores on the Student’s Life Satisfaction Scale (SLSS): r=.55, p<.01; Child and Adolescent Social Support Scale (CASS): r=.53, p<.01; and to a lesser degree the Abbreviated Junior Eysenck Personality Questionnaire (JEPQ-A) Extraversion score: r=.16, p<.05.
The CHS was also negatively associated with the Youth Self-Report (YSR) Externalizing (r=-.32, p<.01) and Internalizing (r=-.32, p<.01 scales), and with the JEPQ-A Neuroticism Scale (r=-.28, p<.01).
3. In a second study, Valle et al. (2004) examined the psychometrics of the CHS with 531 children aged 10-14. Ethnicity was 52% African American, 40% Caucasian, 2% Asian, 1% Hispanic, 1% Native American, and 4% other ethnic background. 58% qualified for a free or reduced lunch. Results were similar to that found in their study of older children. CHS scores were positively correlated with the SLSS (r=.49, p<.01), CASS (r=.59, p<.01), and JEPQ-A (r=.18, p<.01). CHS scores were also negatively correlating with the YSR Externalizing (r=-.33, p<.01) and Internalizing (r=-.32, p<.01)scales.
4. The CHS also correlated significantly with Optimism (r=.54, p<.0001), Pessimism (r=-.34, p<.0001), and Total Optimism (r=.47, p<.0001) scales of the Youth Life Orientation Test (Ey et al., 2005).
5. Soliday, Farofalo, & Rogers (2004) reported a similar correlation between the CHS and the Life Orientation Test (r=.57, p<.001). They also reported significant correlations with the Positive and Negative Affect Schedule for Children, Positive Affect Scale (r=.51, p<.001) and Negative Affect Scale (r=.24, p<.05), Center for Epidemiological Studies Depression Scale (r=-.55, p<.001), Children’s Somatization Inventory (r=-.31, p<.05), and the Youth Self-Report Somatizing Symptoms Subscale (r=-.31, p<.05).
6. Carvajal, Evans, Nash, & Getz (2004) examined the agency and pathways scales separately in a study that examined predictors of adolescent Substance use in 525 adolescents.
Agency was significantly related to: Optimism (r=.60), measured using the Life Orientation Test; Global Positive Expectancies (r=.73); Self-Esteem (r=.58), measured using positive items from the Rosenberg Self-Esteem Scale; Attitudes regarding the negative consequence of substance abuse (r=.37); Subjective norms about using substance abuse (r=.30); Self-efficacy regarding avoiding substances (.33); Intention to use substances (.34); Substance abuse at T1, the number of times the individual used substances in the past 12 months (r=-.27);
and Substance abuse at Time 2 (r=-.234).
Pathways was significantly related to: Optimism (r=.57), Global positive expectancies (r=.80), Self-esteem (r=.55), Attitudes (r=.35), Subjective norms (r=.29), Self-efficacy (.31), Intention (.33), Substance abuse at Time 1 (r=-.26), and Substance abuse at Time 2 (r=-.23). All correlations were significant at p<.01.
DISCRMINANT VALIDITY
1. Snyder et al. (1997): Discriminant validity was examined through correlations with the The Hopelessness Scale, which measures the degree to which children have negative expectancies about themselves and the future. The measures were negatively correlated, but correlations were not statistically significant. The measure was also found to not correlate with verbal, performance, or full-scale IQ scores assessed using the WISC-R or WISC-III.
FACTOR ANALYSIS
1. In addition to the factor analysis conducted during the measure’s development (see “Content Validity”), the authors conducted factor analysis with the five samples of children described under “Population Used to Develop Measure.” The authors reported that overall items loaded on the appropriate factors (Snyder et al., 1997).
2. Valle et al. (2004) used confirmatory factor analytic procedures to test the 2-factor structure proposed by Snyder in a sample of children aged 15-19. Both the Goodness-of-Fit Index (.96) and the Comparative Fit Index (.95) provide support for the 2-factor model, but the Tucker Lewis Index (.87) did not. When compared to a single factor model, the correlated two-factor model provided better fit. They also tested the 2-factor model with a sample of children aged 10-14, with similar results. The 2-factor model again provided significant improvement in model fit over the 1-factor model.
GENDER, AGE, AND RACIAL DIFFERENCES
1. Snyder et al. (1997) reported no statistically significant differences when gender, age, or race was examined (Snyder et. al., 1997). Valle et al. (2004) had similar findings. They found no significant differences for age, and while they found significant differences for gender in one study and race (African American and Caucasian) in both studies, they reported that these differences represented very small effect sizes.
TREATMENT OUTCOME RESEARCH
1. The Children’s CHS has been used in 1 randomized trial, but it did not show significant treatment effects (Soliday et al. 2004).
USE WITH DIVERSE POPULATIONS
1. It appears that the measure was used in a study of Hong Kong Chinese adolescents, but the article could not be obtained (Hui, & Ho, 2004).
TRAUMA-EXPOSED POPULATION
1. The CHS was used in a study by Brown, Houck, Hadley, & Lescano (2005) that examined self-cutting and sexual risk behavior in a group of 293
adolescents receiving intensive psychiatric treatment. Those who engaged in Self-cutting had significantly lower CHS scores than those who did not. A significant number of children had been sexually abused.
| Not Known | Not Found | Nonclinical Samples | Clinical Samples | Diverse Samples | |
|---|---|---|---|---|---|
| Predictive Validity: | Yes | ||||
| Postdictive Validity: |
Snyder et al. (1997) tested the predictive validity of the Children’s Hope Scale with the Iowa Test of Basic Skills in a sample of 372 children (197 boys, 175 girls) in 4th through 6th grades, ages 9-14, attending public schools in Edmund, OK. Scores on the Children’s Hope Scale and the Iowa Test of Basic Skills showed a positive and significant correlation, r(100) = .50, p < .001. CHS scores also contributed additional variance to the prediction of the Iowa scores, above the variance accounted for by the Scholastic Competence
Subscale of the SPP-C and above that accounted for by the Global Self-Worth scale of the SPP-C.
1. Only one study examined agency and pathway components separately (Carvajal et al., 2004). Although there are theoretically two scores, it appears that in general items are summed together to yield a total Hope Score.
2. THIS IS NOT A CON: The author cautions not to overinterpret scores and suggests using a “triangulated” approach for assessing hope by adding supplemental measures such as an
observer rating and behavioral measures.
Translations
| Language: | Translated | Back Translated | Reliable | Good Psychometrics | Similar Factor Structure | Norms Available | Measure Developed for this Group |
|---|---|---|---|---|---|---|---|
| 1. Chinese | Yes |
Population Information
From Snyder et al. (1997)
The measure was originally developed with 372 4th-6th graders (197 boys, 175 girls), ages 9-14 attending public schools in Edmund, Oklahoma. No other demographic information was available on these children.
Psychometrics were examined with 5 other samples of children. Sample 1: 48 boys and 43 girls, aged 8-17 with arthritis, sickle cell anemia, or cancer,
attending a 1-week summer camp held by Children’s Mercy Hospital in Kansas City, Mo. This sample was assessed two times over a 1-week period to examine test-retest reliability.
Sample 2: 170 boys, aged 7-13 with a primary ADHD diagnosis who attended a summer treatment program at Western Psychiatric Inst. and Clinic in Pittsburgh, Pa. Ethnicity data were reported for this sample as: 15% African American, 76.5% Caucasian, and 3% Other. Ethnicity data were missing for 9 children.
Sample 3: 74 non-referred boys w/o an ADHD diagnosis in Pittsburgh, Pa., who were similar in age to Sample 2.
Pros & Cons/References
1. The measure offers information on two components of dispositional hope (Agency and Pathways), although in general, there is only one score for the total measure.
2. Children’s ability and belief in their ability to reach goals is a potentially important domain to assess.
3. Similar reliability and validity data were obtained across samples that included clinical and nonclinical samples and medical samples.
4. The measure is brief (6 items), easy to administer and score.
5. The wording of the items is simple and easy to understand.
6. The measure is easy to obtain.
7. The measure is free.
1. Longitudinal data are not available.
2. Although agency and pathways are separate concepts, the measure does not appear to yield separate scores, or at least the psychometrics were not examined separately for these scales.
3. Research is needed to determine whether the scale can detect change resulting from treatment and examining the use of the measure with trauma-exposed children and adolescents.
A PsychInfo literature search (9/05) of “Children’s Hope Scale” or “CHS” and “Snyder” anywhere revealed that the measure has been referenced in 40 peer-reviewed journal articles. Below is a sampling of these articles.
Note: A number of these articles referenced only the article on the measure’s validation and were focused on the Adult version of the scale or on Hope Theory.
1. Brown, L. K., Houck, C. D., Hadley, W. S., & Lescano, C. M. (2005). Self-cutting and sexual risk among adolescents in intensive psychiatric treatment. Psychiatric Services, 56(2), 216-218.
2. Carvajal, S. C., Evans, R. I., Nash, S. G., & Getz, J. G. (2002). Global positive expectancies of the self and adolescents' substance use avoidance: Testing a social influence mediational model. Journal of Personality, 70(3), 421-442.
3. Ey, S., Hadley, W., Allen, D. N., Palmer, S., Klosky, J., & Deptula, D. et al. (2005). A new measure of children's optimism and pessimism: The youth life orientation test. Journal of Child Psychology & Psychiatry, 46(5), 548-558.
4. Hui, E. K. P., & Ho, D. K. Y. (2004). Forgiveness in the context of developmental guidance: Implementation and evaluation. British Journal of Guidance & Counselling, 32(4), 471-492.
5. Soliday, E., Garofalo, J. P., & Rogers, D. (2004). Expressive writing intervention for adolescents' somatic symptoms and mood. Journal of Clinical Child & Adolescent Psychology, 33(4), 792-801.
6. Valle, M. F., Huebner, E. S., & Suldo, S. M. (2004). Further evaluation of the Children's Hope Scale. Journal of Psychoeducational Assessment, 22(4), 320-337.
The author reviewed this report and had no comments other than that it looks fine as is.