| dc.description.abstract | Despite significant governmental efforts in Kenya to reduce recidivism, the rate of reoffending
among ex-convicts remains high, raising concerns about the effectiveness of existing correctional
programs. This study aimed to examine the effects of imprisonment on recidivism among
recidivist in Kisumu County, Kenya, with specific focus on three variables: prison-based
treatment, inmate interaction, and re-integrative programs. The study was guided by Albert
Bandura’s Social Learning Theory and Howard Becker’s Labeling Theory. This study adopted a
descriptive research design. In addition, the study employed explanatory mixed-methods research
design, integrating both quantitative and qualitative approaches. The target population comprised
946 individuals, from which a sample of 254 was selected using Krejcie and Morgan’s Table and
Neyman’s Allocation Formula. Data collection tools included structured questionnaires and semi
structured interviews. Quantitative data were analyzed using IBM SPSS Statistics for Windows,
Version 25.0. Descriptive statistics (frequencies and percentages) were used to summarize
respondent characteristics and major study variables. Inferential statistics were employed to
examine relationships among the objectives. Specifically, Chi-square tests were used to assess
associations between categorical variables, while Pearson’s Product Moment Correlation
Coefficient measured the strength and direction of relationships between continuous variables.
Additionally, Analysis of Variance (ANOVA) was used to compare group means across key
variables. All tests were conducted at a 95% confidence level, with statistical significance set at p
< .05. Qualitative data, obtained from open-ended survey items and interviews, were transcribed
verbatim, coded, and analyzed using a thematic approach. The research was conducted in three
correctional institutions: Kibos Main Prison, Kibos Medium Prison, and Kisumu Women Prison.
Findings indicated a strong inverse relationship between prison-based treatment and recidivism (r
= .901; R² = .812), and between re-integrative programs and recidivism (r =.-871; R² = .758).
Conversely, inmate interaction showed a significant positive correlation with recidivism (r = .689;
R² = .643). A combined regression model demonstrated that these three variables collectively
explained 66% of the variance in recidivism. The study concludes that imprisonment in its current
form is not uniformly effective in reducing recidivism. Its success depends on the presence of
rehabilitative infrastructure, the management of inmate socialization, and the availability of post
release support. First, the study recommends that correctional institutions invest more substantially
in rehabilitation infrastructure. This includes recruiting qualified trainers and counselors,
upgrading vocational and educational facilities, and ensuring that treatment programs are relevant
to both inmate needs and labor market demands. Second, regarding prisoner interaction, prisons
should design and implement structured engagement programs that foster positive socialization.
Activities such as peer mentorship, faith-based dialogue, recreational sports, and staff-facilitated
group discussions can help transform inmate relationships into tools for behavioral change. Lastly
there is need for establishment of a coordinated framework for post-release support involving
correctional institutions, probation officers, local government, NGOs, and community
organizations. | en_US |
