Forests provide a multitude of ecosystem services to society. However, not all such services are being reflected in market prices and that leads to underestimation of their economic values, and suboptimal management schemes. Therefore, non-market valuation is required to provide complementary information for better forest management that underpin the concept of total economic values. In this PhD thesis, the non-market values of forests are evaluated with a focus to show the impact of forest management on ecosystem services. The thesis consists of four papers that address three main research questions: 1) Which forest structural characteristics and features affect recreational preferences? 2) Does childhood forest experience determine forest visiting habit in adulthood? And 3) How does environmental attitude influence individuals’ willingness to pay for forest management initiatives designed to enhance ecosystem service provision? In the first paper, we evaluate the effects of forest structural characteristics and diversity in forest stands on recreational preferences. The study is undertaken using choice experiment (CE) data about people's preferences for forest characteristics in their future recreational visits. In general, mixed tree species are found to be preferred to monocultures; and stands with varying height are preferred over stands consisting of same height trees. The variation between stands also increases the recreational value of forests; and in some instances, may outweigh the contribution of variation within a stand. In the second paper, we investigate the factors that influence the choice of forest site for recreation. The paper is conducted based on the survey data in which respondents were asked to indicate last visited forest sites using map tools. However, the number of alternative sites accessible to each respondent was too large to include all in the choice set, and hence, we use simple random sampling. Relevant attributes are identified by using spatial data analysis. The factors that significantly influence choice of forest site include: forest area, tree species composition, forest density, availability of historical sites, terrain difference, state ownership, and distance. In addition, we empirically show how large a number of alternatives are sufficient to provide consistent parameter estimates through the random selection. The third paper investigates the impact of past experience, in this case childhood forest experience, on forest visit frequency in adulthood. The study is conducted based on data collected from nine European countries and hence examines cross-country variations in frequency of visit. Childhood experience is found to positively influence forest recreation practices. The frequency of visit is also significantly determined by current residential location and distance (to nearest forest). In the fourth paper, we examine the role of environmental attitude on people’s WTP for forest ecosystem services. We use the new environmental paradigm (NEP) scale to measure environmental attitude as a multidimensional concept. The variation in willingness to pay for ecosystem services is illustrated using different modelling approaches and people with an ecocentric attitude are found to have higher willingness to pay compared to those with anthropocentric attitude. In general, the three main conclusions of the thesis are: o The recreational importance of forests depends on the forest characteristics and other features which in one way or another are influenced by forest management principles. This provides evidence on the necessity to consider the non-market values in forest management so as to maximize social welfare. o Forest recreational behaviour is found to be influenced by individuals’ experiences during childhood, current residential location, and distance. This highlights the importance of improving accessibility of nature areas such as forests for more recreational participation of today’s children – so that they could develop nature recreational habit and pro-environmental behaviour in adulthood. o Individuals’ environmental attitudes on the human-nature relationships influence their WTP for forest ecosystem services. This confirms the importance of understanding the heterogeneity among individuals, before practical policy options are taken from average WTP estimates.