The Study Area and Data Collection The study area is Porsha thana (subdistrict) in the Naogaon district. The total area of Porsha thana is 252.83 km2, and it has a total population of 123,160 where male is 64,760 and female is 58,400 (BBS, 2003). The occupations of the area include agriculture (48.27%), gricultural labor (29.01%), wage labor (3.28%), commerce (5.2%), service (3.6%), and others (10.64%; Banglapedia, 2008). Agroecologically, the study area belongs to High Barind Tract (AEZ-26; Food and Agriculture Organization of the United Nations [FAO], 1988), which is the largest Pleistocene physiographic unit of the Bengal Basin covering an area of about 7,770 km2. It has long been recognized as a unit of old alluvium, which differs from the surrounding floodplains. The sampling unit of this study was each individual household, and a multistage sampling technique was employed to locate these households. Porsha thana consists of a number of villages out of which three were selected randomly. Using the holding tax registers and information gathered in an exploratory phase, 32 households were selected from each village. Thus, in total, 96 households were selected for a detailed botanical survey. To collect household data, slightly modifying the criteria of the BBS (2001b), households were classified into four-landholding size classes— namely marginal (<0.4 ha), small (0.41–1.01 ha), medium (1.02–3.03 ha), and large (3.03 ha). Two methods were employed for data collection: structured questionnaire interviews with the farmers and botanical surveys in the homesteads. The questionnaire was tested in a nearby village before it was finalized. Local community leaders were contacted for cooperation prior to starting the field survey. During the botanical survey, all trees with a diameter at breast height (DBH) >10 cm that were under the possession of the households were identified, and the number of trees in each species were counted. For bamboos, which are typically grown in clusters, the species to which bamboos of a clump belonged were identified and then all matured culms of all clumps were counted and recorded. Theoretical Framework Quantitative stand structural characteristics were analyzed using various stand structure such as relative density (RD), frequency (F), relative frequency (RF), abundance (A), and relative abundance (RA). These characteristics were calculated using the following formulas: RD = (Total number of individuals of a species)/ (Total number of individuals of all species) × 100, F = (Total number of samples in which the species occur)/ (Total number of samples enumerated) × 100, RF = (Frequency of the species in the stand)/ (Sum of the frequencies of all species in the stand) × 100, A = (Total number of individuals of the species in all the samples)/ (Total number of samples in which the species occurred), RA = (Abundance of the species in the stand (Sum of abundances of all species in the stand) × 100. Four non-parametric alpha (α) diversity indices that are often used in studies on species diversity (Magurran, 1988; Krebs, 1989) were used to explain the results that allow opportunities to compare the findings with other studies. These are as follows: 1. Inverse of Simpson’s index (Simpson, 1949; Ito 1997): λ−1 = [ni(ni − 1)/N ∗ (N − 1)]−1, 2. Inverse Berger-Parker dominance index (Magurran, 1988): d−1 = [Nmax/N] −1, 3. Shannon-Wiener index (Odum, 1971; Magurran, 1988): H = −pi ln pi, 4. Evenness index recommended by Pielou (1969): J = H /Hmax ., where, pi is the proportion of the number of i th species (ni) to total number of individuals (N)—i.e., pi = ni/N, and N max = number of individuals of most abundant species, and H max = ln (S), and S = total number of species found in all the samples. Natural logarithm (log scale 2.718) was used during calculation of diversity indices. A beta (β) diversity index—viz., Sørenson similarity index (CS)—was also derived to compare the similarity of species among different landholding size classes using the following equation described by Magurran (1988): Sorenson similarity index Cs = 2j/(a + b), where, j = the number of species found in both sites (A and B), and a = the number of species in site A with b the number of species in site B.