【摘 要】在大多數(shù)社會科學(xué)，研究人員可以使用定量或定性的方法，這是沒有爭議的。與定量研究相比，定性研究帶有更強的文學(xué)風(fēng)格和個人的特點，這是不可避免的，因此定性研究被認(rèn)為傾向于引進研究員的偏見。像其他定性研究方法，扎根理論最反映利用定性研究——調(diào)查過程中避免偏差。扎根理論具有特定程序是靈活的，而且在最寬泛的意義上內(nèi)給予許可信度和效度。

【關(guān)鍵詞】定量研究；定性研究；扎根理論

Quantitative research is always compared with qualitative research， which is the exam， analysis and explanation of researchers for finding implied meanings and combining forms of relationships， including classifications of different kinds of phenomena and activities， in a way that does not include mathematical methods.

In most social sciences， the researcher can use either quantitative or qualitative methods， which is uncontroversial， and each can be used when there is the need. Compared with quantitative research，qualitative research is of more literary style and in more personal voice，which is inevitable to be thought the study which is made under this methology has a tendency to bring in researcher’s bias. As in other qualitative approaches，the grounded theory maybe one of them can most reflect the advantage of qualitative research -- avoiding bias in the process of investigating. Grounded theory has specific procedures and associated canons which is flexible but within limits， which give permit to reliability and validity in the broadest sense. “Formal reliability and validity assessments for grounded theory products， conducted in that order， are consistent with the tenets of this inductive methodology and with measurement theory； they are essential for efficient” （Atwood， Hinds， Benoliel， Artinian，1986） Here are 2 points I got after reading the materials：

1.The concepts we deal with is always “discovered from data （induced），systemically abtained and analyzed”（GlaserStrauss，1967）In grounded theory， analysis is assisted with data collection，the analysis begins as soon as the first a few data is collected.Comparing with other research methods， it’s more scientific，for once the original analysis starts，it can provide materials for the next interview or observations，on one hand，it won’t miss anything since the first data will be analyzed in details.On the other hand， no matter how the investigators are in favor of a particular concept，if it shows no relevence to the topic in the continued study， it must be discarded. The traditional understanding of reliability focuses on standardizing data collection instruments（Mason， 1996） However， this is“premised on the assumption that methods of data generation can be conceptualised as tools， and can be standardised， neutral and non-biased” （Mason， 1996） While this may be acceptable for quantitative methods，but in qualitative research，abviously，the data collection process is more rigorously and patiently treated.In contrast to quantitative research，in which the researcher “systematically applies a pre-existing set of codes to the data”，the qualitative research is “data-drived” （Sandelowski， 2000）.

2.Team work in the grounded theory is an important approach， sharing the experience with colleagues who have experience in the same area gives helps to guard against bias.Our initial concern with subjectivity， reflexivity， and interpretation began a series of questions that took us from the lone researcher to concerns about the methodological issues specific to the work of teams of qualitative researchers （Bresler， Wasser， Hertzog， Lemons， 1996） researchers in a team are not necessarily all agree on the result got from the data. Although we were all collaborators， our roles were not the same， and this difference meant that allegiances and approaches to fieldwork were also asymmetrical （Wasser Bresler， 1996）.After the comparison and analysis， researcher can sift though the usless data and the concepts which are not helpful to the core concept. Team work plays an nonnegligible role in qualitative research that it’s necessary for the group to have a periodic discussion or data exchange.

So these are two procedures in the grounded theory that can most avoid being “over-subjective” and bringing in the researcher’s bias， the concepts won’t deviate from the original purpose， and they should be skillfully used in our future research.