Why do you want to move up a level? For most amateur players the answer is to improve their games; play against better competition, which will be both more challenging and more educational. For the aspiring pro, however, the answer is far more financially motivated. The higher the game, the more money can be made. If poker is what pays your bills, then winning at a higher level generally means more money. This is not to say that either group doesn't care about the other consideration. Amateur players like money, and pro players care about the challenge of poker and improving their games too. Nevertheless, each group has a different focus and motivation, and I would argue that even within the groups you would find a wide variety of motivations.
Despite these differences, the vast majority of advice available online is geared toward professionals and semi-professionals. The key idea is not to go broke, and this makes a lot of sense for professional players. If your bankroll disappears, then you no longer have the means of doing your job. Amateurs, however, can reload with the next paycheck from their "real" jobs. Risk of ruin, i.e., the risk of losing your entire bankroll, is used in many of the formulas one finds to calculate bankroll requirements, but this idea has limited use for the amateur player, and even for the professional. If you want to avoid going bust, all you need to do is move down in levels and reduce the risk, assuming you are a winning player. Risk of ruin just means if you continue to play at a particular level, you have X% chance of losing your entire bankroll; thus if you are willing to move down levels, you can effectively make your risk of ruin almost nothing. For the professional, however, moving down a level is a significant cut in income and so is not something one wants to do. This is why there is so much focus on protecting the bankroll. Amateurs, however, are playing to improve their games, not to make a living. Moving down a level is easy for the amateur. This is not to say that amateurs can ignore bankroll. Even if you are capable of replenishing your bankroll, you may not want to. If you have a sizable enough bankroll to play .50/1 NL, you may not have that amount of cash just sitting around waiting to be inserted as a new poker bankroll. If you are fortunate enough to be in a position in which money is no object, call me; let's be friends.
Determining whether you are ready for the next level requires that you have some figures on hand. You will need your win rate and the standard deviation of that win rate. Win rate is usually expressed in hourly terms ($10/hour) or in big blinds per 100 hands (5BB/100). Limit games generally express win rate in terms of big bets, which also uses the BB/100 notation, and some people express their NL win rates as big bets (which is just twice the big blind); hence it can be confusing, but for present purposes if I write BB/100, I mean big blinds per 100 hands. Professional players often express win rates in hourly terms and understandably so. As a professional, it would be good to know if you are making more per hour than minimum wage. For the amateur, however, poker is not a job and determining whether you are a winning player is generally expressed in terms of BB/100, and that is the expression I will use here. In order to get these figures you need to keep accurate records. Tracking programs such as Poker Tracker are designed for this purpose. You can keep track yourself, though likely with less specificity and breadth. At a minimum you need to record the number of hands you play per session and the amount you win or lose in each of those sessions. Standard deviation is calculated for you by most tracking programs. If you track your own results in a spreadsheet such as Excel, then calculating standard deviation can be done without much difficulty.
In order to even consider moving up, you need to be a winning player at your current level. To know whether or not you are a winning player, you need to have a significant enough sample size. 10,000 hands are generally considered to be the minimum sample. I think you can probably get away with less if you are absolutely crushing a game. For example, if you are crushing a NL game to the tune of 20BB/100 over 6000 hands, you are most likely a winning player. With stats that good, you really don't need to play 4000 more hands to know, with reasonable certainty, that you are a winning player. Your win rate, however, may be significantly inflated if only 6000 hands are under consideration. You may be a winning player, but just not as good as your current win rate suggests. The more accurate your win rate is, the better idea you will have regarding how much bankroll you need to move up. Conversely, if you are only winning 2BB/100 over 6000 hands at NL, then you really need to put more hands in to know if you are a winning player, or have just been lucky, or perhaps unlucky with a true win rate that is much higher.
Skill wise, the higher your win rate the better, of course. I think you can consider moving up a level if you have a 3BB/100 rate and the proper bankroll. I would not expect to duplicate your win rate at the next level, though. The competition will be better, and this means that your mistakes will cost you more, because your opponents will be more likely to take advantage of these mistakes than your current opponents.
Ideally, one wants at least a 5BB/100 rate. This is a solid win rate, and you are basically ready for the next level of competition; all you need is the proper bankroll. Moving up with only a 3BB/100 rate may put you in a breakeven situation at the next level, but you can consider that to be payment for the "lessons" you will receive. The better competition will improve your game, but it will be trial by fire. If that sounds highly unpleasant to you, then stay at your current level until you improve your game.
It is also preferable to collect data from the same site. Play at Ultimate Bet is much different from play at Bodog . Each site has its tendencies and usual players. You may do well at one site and not so well at another. Hence your BB/100 rate at Ultimate Bet is not a reliable estimate for how you will do at Bodog and vice-versa. Even if you have similar win rates at two sites, you may experience greater variance at one leading to vastly different standard deviation rates (and this will matter when we get to bankroll considerations in Part II). Thus figure your stats by site and consider moving up on a particular site; not on all sites simply because you are moving up on one of them.
Even if you decide that your game still needs work before moving up, I recommend taking some shots at the next level. It really will help you improve. Consider the following analogy. You try to improve your basketball game and you play in a regular weekly 5 on 5 with other out of shape people on the weekend. You get a chance to play with the local NCAA collegiate team. (Ignore why and how such a chance came along). How much more quickly will your game improve playing with people who "know" what they are doing? The same applies in poker. Better opponents will show you things that lesser ones just can't. Thus, while you want to be a winning player before moving up a level, moving up a level will also make you a better player.
Next time I'll examine bankroll requirements for moving up a level. I think traditional requirements are inexact, leading some to stay at a current level for too long and others to move up too soon. Moreover, though more precise measures exist, these are geared for professional players rather than the amateur.