This study looks at the effectiveness of direct and indirect written corrective feedback (WCF) when using them with 26 intermediate ESL learners’ writings. The study also investigates whether giving the learners the type of feedback they preferred or less preferred would influence their actual performance. WCF, a pedagogy that is often used when helping learners improve their written accuracy, has captured researchers’ attention in recent decades. Truscott (1996) claimed that WCF is ineffective and suggested teachers should abandon it. Therefore, in the early studies, researchers concentrated on examining the effectiveness of WCF, in order to justify the value of using WCF. In recent studies, researchers have proved that WCF is effective in certain contexts, and they have also investigated the value of using different types of WCF, and the value of using it over time. Moreover, in order to help learners to improve in written accuracy, recent studies in the field have also investigated whether WCF is more effective when used on a certain type of linguistic form/structure. With respect to different types of WCF, researchers in recent decades have also paid some attention to learner preference in WCF. However, the relationship between learner preference and the value of using the type of feedback learners prefer has not yet been investigated. In essence, the relationship between learner preference and their actual performance when using the type of feedback they preferred was examined in this study. The study also aimed to look at the effectiveness of WCF over time, and to investigate whether direct feedback or indirect feedback helped learners better. Furthermore, the study also aimed to investigate whether there was a certain type of linguistic form that WCF works best with. A quantitative approach was used in this study in order to show the results more clearly, and to provide statistical evidence on each finding. The study involved questionnaires, and three writing tests: pre-test, immediate post-test, and a delayed post-test. Before the learners did the pre-test, they were asked to complete a questionnaire to select their preferred type of feedback (direct feedback or indirect feedback). Based on their preferences on feedback, they were put into different groups. Group one received direct feedback; group two received indirect feedback; group three received the feedback they preferred (indirect feedback); group four received no linguistic feedback, but general commentaries on their writing were given. The participants (twenty-six students who enrolled in a general English program at AUT University) had completed the questionnaire and the three writing tests. Findings from the study revealed that, most of the learners preferred receiving direct feedback. When examining learner performance between those who received the type of feedback they preferred and those who did not, the former did not outperform than others. However, factors like different levels of scaffolding assistance may have affected the results. Other findings from the study revealed that WCF was effective overtime, especially when using it on errors of present simple tense. The study also found that indirect feedback was more effective than direct feedback. A possible factor that appeared to influence learners’ performance was identified as learners’ motivation in learning. The results of the study contribute to an understanding of the type of feedback that is most suitable for learners at intermediate level, and on which type of linguistic form WCF can work best with. Practical suggestions for pedagogy and further research are also made.