This paper assesses the use of survey research in Management Information Systems (MIS) from 1980 to 1990. It identifies five major weaknesses in the application of survey research in MIS: (1) single method designs where multiple methods are needed, (2) unsystematic and often inadequate sampling procedures, (3) low response rates, (4) weak linkages between units of analysis and respondents, and (5) over reliance on cross-sectional surveys where longitudinal surveys are really needed. The paper also shows that the quality of survey research varies considerably among studies of different purposes: explanatory studies are of good quality overall, exploratory and descriptive studies are of moderate to poor quality. The paper presents a general framework for classifying and examining survey research and uses this framework to assess, review and critique the usage of survey research conducted in the past decade in the MIS field. In an effort to improve the quality of survey research, this article makes specific recommendations that directly address the major problems highlighted in the review. Survey research is a quantitative method that requires standardized information from and/or about the subjects being studied. The subjects studied might be individuals, groups, organizations or communities; they also might be projects, applications, or systems. The main way of collecting information is by asking people structured and predefined questions. Their answers, which might refer to themselves or some other unit of analysis, constitute the data to be analyzed. The nature of survey research can be best understood by comparing it to two other dominant methods in MIS: case studies and laboratory experiments. Case studies involve examination of a phenomenon in its natural setting. The researcher has no control over the phenomenon, but can control the scope and time of the examination. The researcher may or may not have clearly defined independent and dependent variables. Case studies are most appropriate when the researcher is interested in the relation between context and the phenomenon of interest. Laboratory experiments involve examination of a phenomenon in a controlled setting. The researcher manipulates the independent variables and observes their effects on the dependent variables. The researcher has direct control over the laboratory conditions and manipulation of the independent variables, but the researcher can only study phenomena in the present. In contrast to these two methods, survey research involves examination of a phenomenon in a wide variety of natural settings. The researcher has very clearly defined independent and dependent variables and a specific model of the expected relationships which is tested against observations of the phenomenon. Survey research is most appropriate when: (a) the central questions of interest about the phenomena are "what is happening?", and "how and why is it happening?" Survey research is especially well-suited for answering questions about what, how much and how many, and to a greater extent than is commonly understood, questions about how and why. (b) Control of the independent and dependent variables is not possible or not desirable. (c) The phenomena of interest must be studied in its natural setting. (d) The phenomena of interest occur in current time or the recentThis paper assesses the use of survey research in Management Information Systems (MIS) from 1980 to 1990. It identifies five major weaknesses in the application of survey research in MIS: (1) single method designs where multiple methods are needed, (2) unsystematic and often inadequate sampling procedures, (3) low response rates, (4) weak linkages between units of analysis and respondents, and (5) over reliance on cross-sectional surveys where longitudinal surveys are really needed. The paper also shows that the quality of survey research varies considerably among studies of different purposes: explanatory studies are of good quality overall, exploratory and descriptive studies are of moderate to poor quality. The paper presents a general framework for classifying and examining survey research and uses this framework to assess, review and critique the usage of survey research conducted in the past decade in the MIS field. In an effort to improve the quality of survey research, this article makes specific recommendations that directly address the major problems highlighted in the review. Survey research is a quantitative method that requires standardized information from and/or about the subjects being studied. The subjects studied might be individuals, groups, organizations or communities; they also might be projects, applications, or systems. The main way of collecting information is by asking people structured and predefined questions. Their answers, which might refer to themselves or some other unit of analysis, constitute the data to be analyzed. The nature of survey research can be best understood by comparing it to two other dominant methods in MIS: case studies and laboratory experiments. Case studies involve examination of a phenomenon in its natural setting. The researcher has no control over the phenomenon, but can control the scope and time of the examination. The researcher may or may not have clearly defined independent and dependent variables. Case studies are most appropriate when the researcher is interested in the relation between context and the phenomenon of interest. Laboratory experiments involve examination of a phenomenon in a controlled setting. The researcher manipulates the independent variables and observes their effects on the dependent variables. The researcher has direct control over the laboratory conditions and manipulation of the independent variables, but the researcher can only study phenomena in the present. In contrast to these two methods, survey research involves examination of a phenomenon in a wide variety of natural settings. The researcher has very clearly defined independent and dependent variables and a specific model of the expected relationships which is tested against observations of the phenomenon. Survey research is most appropriate when: (a) the central questions of interest about the phenomena are "what is happening?", and "how and why is it happening?" Survey research is especially well-suited for answering questions about what, how much and how many, and to a greater extent than is commonly understood, questions about how and why. (b) Control of the independent and dependent variables is not possible or not desirable. (c) The phenomena of interest must be studied in its natural setting. (d) The phenomena of interest occur in current time or the recent