Investigation of the Factor Structure of Mentoring Relationship Quality in Poor Children of Mentorship Program from a Hong Kong Asset-Based Child-Development Program

Background and Purpose: This study identified the factor structure of mentoring relationship quality (MRQ) items adapted from four commonly used mentoring scales in western mentorship programs. The four mentoring scales showed good psychometric properties in the western programs and were adapted for an evaluation study of the first batch of the on-going territory-wide government-funded Child Development Fund (CDF) projects in Hong Kong which aims to build financial and non-financial assets in children aged 10-16 from financially disadvantaged families. Mentorship program, as one of the three components of the CDF projects, contributed uniquely in the development of the participants and understanding of MRQ among Hong Kong Chinese children of disadvantaged backgrounds, and comparison with existing western understandings, helps to inform the building of good mentoring relationship and facilitate the success of mentorship program in Hong Kong.

Methods: All participants (n = 750) of the first batch CDF projects were invited to participate in the evaluation study using longitudinal survey design. Thirty-one items assessing MRQ adapted from four mentoring scales with good psychometric properties were included in the questionnaires and data from the middle three rounds were used because MRQ was not assessed in the first round before the mentor-mentee matched and the last round survey showed poor response rate.

A two-step approach of factor analysis was performed for the longitudinal data of MRQ. Exploratory factor analysis (EFA) was first step performed on the second round MRQ data to identify a stable factor structure, and the second step we used confirmatory factor analysis (CFA) to confirm the identified factor structure on the third and fourth round MRQ data.

Results: A seven-factor structure on the 31 MRQ items was identified in the EFA and confirmed in the CFAs. The seven factors are youth-centered relationship, positive emotional engagement, no negative emotional engagement, trust, psychological proximity seeking, help to cope, and empowerment and performance standard. Chi-square values for the factor structure in CFA were all significant (p < 0.01) but the goodness-of-fit indices showed good representation of the MRQ data with the identified structure (CFIs ranged from .93 to .95; RMSEAs ranged from .05 to .06; SRMR ranged from .03 to .04). Internal consistency reliability coefficients (Cronbach’s α) of the factor items were all over .80 except for the factor of trust, which consisted only two items.

Conclusions and Implications: Factors identified from the MRQ items adapted from existing validated mentoring scales in from mentorship programs for poor Hong Kong Chinese children are similar to those found in the western mentorship programs. The factors also follow the understanding of MRQ with two core dimensions, namely relational and instrumental, in the mentoring relationship literature. Mentorship program has a history of over a century in the US and it is blooming in Hong Kong under the support of the government. Social workers responsible for operating the programs and managing mentoring relationship can benefit from the findings in the training of mentors as well as facilitating mentor-mentee relationships.