Sociological research, the attempt to investigate human society by observing, and measuring the behavior of groups within that society is fraught with problems, not least the problem of objectivity. No matter how much scientific method is applied to these studies, or how refined the tools of measurement become, the observer will always be a human being, and as such a part of the society being studied.
Nevertheless a good introduction to statistics is essential for the potential social scientist, although it is debatable how accurate statistical analysis can ever be. Measuring probability based on available data, identifying trends and attempting to predict human behavior can never account for the vagaries of the human imagination, let alone explain behavior based on that imagination. Science observes the observable; and the human mind, not to be confused with the brain, has yet to be dissected for scientific scrutiny.
With these shortcomings in mind, different ways of sociological research are available. Participant observation is certainly a useful method, but again its peculiar problems must be understood.
Joining a group to observe its behavior from within can provide many useful insights, but yet this is a method with huge inherent difficulties. Being a secret member of a group smacks of spying, the mole within; recording observations as they occur might be impossible without giving the game away; and becoming part of a group for any length of time does tend to bias the sympathies of the observer.
As an attempt to unravel the dynamics of society longitudinal surveys can be especially useful. Examine a group and return after a period of time to monitor any changes. The questionnaire, a long favored tool of social scientists, is valuable in this type of study. Repeat the original questions after a period of time and see how the responses differ. This can be a useful tool for recording changes of attitudes.
A further cautionary note is needed. Attitudes are not exactly observable and are frequently misleading if not downright wrong. If any precision could be attributed to them, election results would always be known in advance. Despite the millions of dollars spent on pre-election surveys (i.e. questionnaires) common sense and experience suggests that they are miserably inaccurate predictors.
Of course it is notoriously difficult to identify a group. A member of a family is frequently a worker, be it white or blue collar, with political and religious allegiances that are not always readily definable, or even easy to understand. Any sociological research has to account for the fact that most people are simultaneously members of widely different social milieus; with attitudes and beliefs that don’t always conform to the rigid parameters of any one specific faction.
Never forget the importance of library studies. It is sometimes a conceit of sociologists to imagine that their viewpoint provides the only valid record of contemporary society but quality journalism does, after all, endeavor to do the same job as sociology. It observes contemporary society in action and usually attempts to differentiate between fact and opinion.
Fiction too, if it reflects any truth at all, has a lot to say about the state of the world.
Genre writers, especially in crime fiction, often reflect the dynamics of today’s society. Walter Mosley’s series of novels, featuring hard boiled detective Easy Rawlins, springs readily to mind. Ostensibly a writer of crime fiction, his books contain lots of relevant comments regarding the black experience in America since World War 2. These observations may, or may not, be more accurate than sociological ones, but they do have the benefit of being easier to read than much of the often impenetrable prose found in many sociological journals.
Methodology should be as free of experimenter bias as possible; but experimenters can never escape the fact they too are members of the different sectors of the society that they study. An unkind comment on social scientists is that they are merely looking out of the window; budding, and even practicing, social scientists might do well to bear that in mind before making any extravagant claims for the validity of their findings.