Population and Sample To exclude other people when a comment is intended for only a limited audience

30 c. Identifying the words or sentences in the conversation which belong to code switching and code mixing. d. Classifying and analyzing the types of code switching and the forms of code mixing. e. Calculating each type of code mixing and determining the most dominant type. f. Drawing some conclusions based on the result of analysis. The method of study applied in this thesis is field research. The writer uses the methods to find out the data and relevant information that will support the analysis by collecting the sampling and data by making recordings of the conversation or utterances and finishing the analysis based on the real fact directly.

3.1 Population and Sample

To complete this research, the writer needs population and sample. The population in this thesis is some of the admin staffs of CaSEducation that can be recognized to be able to speak English well. Sometimes taking sample in qualitative research can be more and less, the information which is gotten can be maximal, like the statement of Patton 1975 in Guba and Lincoln, 1985:233 “The sample is to be selected in ways that will provide the broadest information possible. Thus, what is meant by a qualitative isomorphs in this case is a qualitative informational isomorphs, that is a sample that is expanded until redundancy with respect to information is reached at which point sampling is Universitas Sumatera Utara 31 terminated. Thus sample maybe large or small, but it is sufficient when the amount of new information provided per unit of added resources expenditure has reach the point of diminishing returns that is it would not be profitable to add even one more sample element” Purposive sampling Lincoln and Guba, 1985:201 – 202; Lexy Moleong, 2000: 165 – 166 can be characterized as follows: a. Emergent sampling design, the sample can not be found or taken first b. Serial selection of sample units, if the purpose is to maximize information, the sampling is terminated when no new information is forthcoming from newly sample units, thus redundancy is primary criterion. c. Continuous adjustment, get more information can expand a hypothesis based on the focus of research. d. Selection to the point of redundancy, selecting the last sample if there is no more information can be found. The writer chooses staffs that are considered will give the data that the writer need. After that, from those samples, the samples, the writer will choose more people to be the samples to give more complete data. But, in purposive sampling, samples are taken by information consideration. Further, Bouma Gary D 1993:119 in his book “The Research Process” revision edition says “Purposive sampling. Some researchers believe that they can using judgment or intuition, select the best people or groups to be studied. The typical rural school is selected and studied, and the results generalized t o rural school”. Universitas Sumatera Utara 32 From the quotation above, the writer has to find out the typical and the best staffs who can give the complete data. William, at. Al 1982:107 says “Respondent who are hard to hard to locate and crucial to the study”. They are wished can give tha unique experience and the knowledge or the data about code switching and code mixing that is needed by the writer.

3.2 Location and Time of Observation